A study on the attitudes towards generic drug
in the Netherlands
Erasmus Universiteit Rotterdam
Erasmus School of Economics
Department of Marketing
Supervisor: dr. Isabel Verniers
Stefan van der Goes
315306
31 July 2011
MASTER
THESIS GENERIC DRUGS: THE
DUTCH POINT OF VIEW
Preface
This thesis marks the end of my master‟s programme for my study Marketing at the Erasmus
University Rotterdam.
It is the result of months of work, and is has been an interesting and informative process.
However, I could not have finished this thesis without the help of certain people.
Therefore, I would like to thank them for their support. First of all, I would like to thank my
supervisor, dr. Isabel Verniers, for her advice, motivation and support.
Secondly, I would like to thank my brother Michel, for helping me out with questions
concerning English language and grammar.
Lastly, I would also thank all the respondents for taking the time to fill out my questionnaire.
Naaldwijk, July 2011
Stefan van der Goes
Source picture frontpage: http://www.geninv.net/2010/04/does-the-new-health-care-bill-change-
access-to-generic-drugs-in-the-us-medicine/
Execute Summary
All western countries have faced the same problem over the past years: How to stop the rise
of public healthcare costs? The aging of the society and the fact that elderly people live
longer were two causes for this rise.
Generics were seen as a solution to this problem. Generics are made when a patent of a
brand name drug expires. These generics are 30 to 70% cheaper than the brand name
drugs. The Dutch government implemented a policy in 2008 which implied that only the
cheapest drugs would be reimbursed (exceptions vary by insurer).
This study aims to find out if there are differences between people with certain demographics
and their attitudes towards generics. Furthermore, it aims to provide insight into the
relationship between shopping behavior and attitudes towards generics.
Not many differences between demographics and attitudes towards generics were found.
However, one finding was that people in an urban environment have more negative attitudes
towards generics compared to people in semi-rural or rural environments.
For shopping behavior a factor analysis was conducted. It was found that 4 factors
determined attitudes towards generics; awareness, (perceived) quality, price sensitivity and
status. In a linear multiple regression, only price sensitivity was significant.
A linear multiple regression of all variables used in this study showed that price sensitivity
had a positive influence on attitudes towards generics and urban environment a negative
influence.
As pharmacies do have own brands (generics) too, it is especially advised for those in urban
environments to communicate the low(er) price saving more clearly. Only a low price can
contribute to a more positive attitude towards generics.
Furthermore, this study provides solutions for brand name drug companies on how to
compete with generic manufacturers.
One solution is to be innovative. By innovativeness, brand name drug companies can
differentiate themselves from generic competitors. Moreover, by being innovative the chance
to find a new drug is bigger, which gives a firm a patent for several years.
Another way to compete with generic manufacturers is to focus on the prescribers of drugs,
i.e. general practitioners and pharmacists. They are sensitive to promotions of a drug and will
prescribe a drug on promotion more than its generic competitor.
The last option is to collaborate with generic manufacturers. This could be done via
ingredient branding. This means that a brand name manufacturer will provide the main
ingredient for a generic drug (as a sign of quality). If this ingredient branding does not lead to
a higher price for the generic drug, both the brand name manufacturer (a bigger market) and
the generic manufacturer (a higher perceived quality) will benefit from this collaboration.
Table of Contents
1. Introduction 7
1.1 Generic drugs: the same, but different 7
1.2 Situation abroad 8
1.3 Situation in the Netherlands 69
2. Theoretical Background 11
2.1 Dutch policy on generic drugs 14
2.2 Generics and Private Labels 15
2.3 Attitudes and trust towards generics 19
2.4 Hypotheses 21
2.4.1 Gender 21
2.4.2 Age 22
2.4.3 Type of environment 23
2.4.4 Knowledge about generics 25
2.4.5 Use of generics 25
2.4.6 Social Economic Status 26
2.4.7 Type of disease 27
2.4.8 Brand Loyalty 28
2.4.9 Price, (perceived) Quality and Status 29
3.Methodology 31
4.Results 33
4.1 Data exploration 33
4.2 Gender 34
4.3 Age 34
4.4 Type of environment 35
4.5 Knowledge about generics 36
4.6 Use of generics 36
4.7 Social Economic Stats 37
4.8 Type of disease 38
4.9 Brand Loyalty, Price, Quality, and Status 38
4.10 Private labels 32
5. Discussion 46
5.1 Demographics 46
5.2 Behavioral factors 46
5.3 Recommendations for generic drug manufacturers 47
5.4 Recommendations for brand name drug manufacturers 47
5.5 Limitations and future research 48
6. References 51
7. Appendices 58
Appendix A: Demographics 58
Appendix B: Questionnaire 59
Appendix C: Overview of urban, semi-rural and rural 67
Appendix D: Policy regarding generic drugs (in Dutch) 70
Appendix E: Output hypothesis 1 72
Appendix F: Output hypothesis 2 73
Appendix G: Output hypothesis 3 74
Appendix H: Output hypothesis 4a 75
Appendix I: Output hypothesis 4b 76
Appendix J: Output hypothesis 4c 77
Appendix K: Output hypothesis 5 79
Appendix L: Output hypothesis 6a, 6b, 6c and 6d 81
Appendix M: Output overall regression 87
Appendix N: Output factor analysis private labels 92
7
Master Thesis Stefan van der Goes
1. Introduction
As expenses for public health in almost every western country rose over the past 10 years
(Organisation for Economic Cooperation and Development [OECD], 2011) new policy was
needed in order to control these costs. Continuing with the old policy, the fear was that costs
for public health would rise even further (OECD, 2006). More and more countries view
generic drugs as the solution for this problem, including the Netherlands. But what do the
people in the Netherlands think about generics? Will they accept a drug which is different,
but in a way the same as their brand name drugs? Which people will? Which people won‟t?
This study aims to find answers to these questions.
This chapter first will provide general information about generic drugs and their
manufacturers. Next, examples of foreign countries who already implemented changes in
their policies in favour of generic drugs are examined. Lastly, the Dutch situation concerning
costs of public health and generics is discussed.
1.1 Generic drugs: the same, but different
Generic drugs are seen as a solution for the problem of the rising healthcare costs. Generic
drugs (generics) are „kind of‟ the same drugs, but at the same time different. When a
pharmaceutical company has, after years of research and testing, found a new drug, this
drug (brand name drug) is patented for several years1. These patents are of vital importance
to the pharmaceutical companies. In this way the company can earn back their research
costs, make some profits and can invest in research.
This monopoly due to the patent is in line with Schumpeterian economics. Schumpeter said
that „technological innovation often creates temporary monopolies, allowing abnormal profits
that would soon be competed away by rivals and imitators. (…) These temporary monopolies
are necessary to provide the incentive necessary for firms to develop new products and
processes’ (Pol & Carroll, 2006).
When a patent is expired, other companies are free to copy this drug. These companies
didn‟t have to incur the costs of years of research, and therefore their price will be (much)
lower than the brand name drug. But before this generic can come to the market, is has to
meet the same quality requirements as a brand name drug, the working ingredient must be
the same as in the brand name drug and the doses must be the same.
1 This patent differs by country.
8
Master Thesis Stefan van der Goes
Generics differ from brand name drugs (besides being cheaper), because they look different,
and come in a different package. So, patients may be a bit reserved when offered a generic,
because they don‟t recognize it and therefore may not trust it.
Besides patients that may be averse towards generics, pharmacies are not always keen on
generics. Pharmacies sometimes loose income by the elimination of discounts and bonus
plans by brand name drug companies (Mott and Cline, 2001). On the other hand, they could
make more profits by selling their own brands (i.e. private labels for cough drops).
Generic manufacturers are booming. They are competing with brand name drug companies
or collaborating with brand name drug companies. For instance, AstraZeneca and Teva,
Teva will not bring generics on the market of AstraZeneca‟s Nexium until 2014 (White, 2010).
A new trend is that generic companies start to launch new brand name drugs themselves
(Goldstein, 2010).
1.2 Situation Abroad
In the period of 2000 – 2009 the consumption of prescribed and non-prescribed drugs both
rose (CBS, 2009). This trend is not unique for the Netherlands, most European countries and
the USA showed a similar trend. Compared to other Europeans, the Dutch use less drugs2.
This however doesn‟t stop the costs for health expenses to grow in the Netherlands.
European countries and the USA all have this problem, caused by the babyboom after WW
II. Some countries have taken action to stop the rise in health expenses.
Australia was one of the first countries where rules concerning reimbursement of cheaper
drugs were implemented. In 1990 the Minimum Pricing Policy was introduced, but its impact
was small. In 1994 legislation involving generic substitution was accepted. In the first 5
years, the use of generics rose from 17 to 45% (McManus, Birkett, Dudley and Stevens,
2001).
Generic substitution in Germany became active in the 1992. Total savings that resulted from
this substitution in the 1990s were 425 million D-Mark (€ 217,30 million) (Schneeweissab,
Schöffskic and Selked, 1998).
In March 2003, new pharmacy legislation was valid in Norway. It contained the introduction
of substitution of generics for brand name drugs. This measure was taken primarily to reduce
costs (Kjoenniksen, Lindbaek and Granas, 2006). It was allowed for patients and/or general
2 http://www.rijksoverheid.nl/onderwerpen/geneesmiddelen retrieved on 20-05-2011
9
Master Thesis Stefan van der Goes
practitioners to refuse generic drugs, but as a result they had to pay a higher price for the
brand name drugs.
Another Scandinavian country, Sweden, had taken action to stop the increase in costs for
public health as well. Sweden used to have the policy that the state insurance company
(most of the time) reimbursed one‟s expenditures due to prescribed drugs. In 2002 a similar
system as in Norway was introduced (Andersson, Sonesson, Petzold, Carlsten & Lönnroth,
2005). Five years after the introduction of the new system of reimbursement, 40% of the
people who switched to generics claimed to have experienced at least one difficulty related
to the use of generics. (Frisk, Rydberg, Carlsten & Ekedahl, 2011).
In Finland a similar study was conducted 5 years after the introduction of generic
substitution. A large majority (70,9%) was satisfied with the switch (Heikkilä, Mäntyselka and
Ahonen, 2010).
Norway, Sweden and Finland all saw a decline in costs for drugs after the new policy.
Furthermore, the majority of the patients in all three the countries were satisfied with the new
legislation (Andersson, et al., 2005; Frisk, et al., 2011; Kjoenniksen, et al., 2006).
The situation in the USA is more difficult to compare, because the rules of reimbursement
differ per insurance company and the reimbursements by the state differ greatly from
European countries. However, from 1999 till 2008, the use of generics saved the USA $734
billion (Vaitheeswaran, 2009).
1.3 Situation in the Netherlands
The Netherlands are no exception compared to other countries regarding to the rise in costs
of public health. Current expenditures on public health in 2009 amounted € 9,585 billion,
which is almost three times as high as in 2000 (Central Bureau of Statistics [CBS], 2011).
A growing trend was visible concerning drug usage. For prescribed drugs the rise was 6%
(34 to 40%), for non-prescribed drugs the rise was 4% (35 to 39%) (CBS, 2011).
This increase means that the expenditures for healthcare (looking only at drugs) are growing
too.
Not only the use of more drugs is a cause of more expenses, the aging of the country is
another factor that plays a role of the rise of more use of drugs. This aging is due to a growth
in elderly people, and this elderly people are living longer too3 (CBS, 2011).
3 2011 is a milestone in the aging of the population. This year the first baby boomers will become 65 years. In five years from
now, there will be half a million people who will be 65 or older. This growth is twice as much as the past five years (CBS, 2001)
10
Master Thesis Stefan van der Goes
A policy was accepted in 2008 that insurance companies only reimbursed the cheapest drug
(exceptions vary by insurer).
In order to make this generic substitution work, it is important for generic manufacturers,
insurance companies and governments that patients really trust generics. In this study I
would like to explore the attitudes4 of Dutch people towards generics. Different parties may
benefit from a more understanding of these attitudes:
generic manufactures, insurance companies and policy makers because they will get
more insight into who they have to target, in order to measure to make that those
people will accept generics too;
brand name drug manufacturers because they will get more knowledge what kind of
people do actually prefer brand name drugs and can market on them;
brand name drug manufacturers because they will get more insight to how they can
compete with generic drug firms.
As different people have different opinions, I will segment my research group. It is interesting
to explore the differences between gender, age groups, urban/rural people, people of
different Social Economic Statuses, and people with different knowledge of generics.
Furthermore, preferences concerning brand loyalty, price sensitivity, perceived quality and
status will be studied, because these can influences people‟s attitudes towards generics.
Information from different Dutch agencies was used to examine the current situation
concerning public health, policy towards generics, fact and figures about the Dutch
population, etc. The names of the authorities are translated in English. In case it was
necessary, the Dutch name was mentioned italic in brackets.
4 There are several definitions, however I choose for the definition given by Oppenheim (1992): An attitude is a state of
readiness, a tendency to respond in a certain manner when confronted with certain stimuli. They formulated this definition after researching several articles about attitudes.
11
Master Thesis Stefan van der Goes
2. Theoretical background
So far, there hasn‟t been many studies that examined the attitudes towards generic drugs.
Moreover, not a single study was conducted in the Netherlands.
In this chapter there will nevertheless be an overview of research that is similar to my topic.
Most comparable to this topic, was the study done by Kjoenniksen, et al. (2006). Their topic
was Patients’ attitudes towards and experiences of generic drug substitution in Norway.
The results of their study were that about 2/3 of the patients claimed overall satisfaction after
the substitution to generics, but about 1/3 of the patients who had their medication
substituted reported negative experiences. 36% Of this group had one or more negative
experiences, 21% claimed an overall negative experience due to the switch to generics. For
patients who reported one or more, or overall negative experiences, generic drug substitution
is not an equal alternative to branded drugs. A solution is to give those patients extra
information and support about generics.
The results are independent of age, gender, number of medications or the physician.
Suh (1999) has researched the trends of generic substitution in community pharmacies in the
United States. The pharmaceutical purchases (of brand name drugs and generics),
represented in 1999 approximately 6% of all pharmaceutical purchases. There is high growth
in the purchase of generics, and several reasons are named to cause this effect, namely:
there is more availability of generic substitute, the generic substitution rate is increasing and
the fast availability of a generic drugs after patent experience. Furthermore, Suh found that
25% of the physicians still refuse to prescribe generics. Pharmacists aren‟t always keen on
generics as well. Many of them are concerned about the frequent switching of generic
manufacturers (Suh).
Moreover, Suh found that generics are at market-entry about 75% of the price of a brand-
name drug. Within one year, this is dropped to 46%, and it is 25% of price of the brand name
drug in three years. So, from a patient point of view, there is a financial incentive to choose
generics instead of brand name drugs.
The main conclusion of this article is that generic substitution can be maximal when using a
multidisciplinary team approach, in which all parties (general practitioners, pharmacists and
insurance companies) involved in the drug treatment process are present.
Research of Andersson, et al. (2005) focused on the Swedish situation. In Sweden
mandatory generic substitution was introduced in 2002, mainly to reduce the increasing
12
Master Thesis Stefan van der Goes
pharmaceutical expenditure. This was necessary because over a period of 10 years the total
pharmaceutical expenditure had become twice as high.
They pointed out that pharmacy personnel is an important factor in generic substitution,
because „they influence the outcome of the reform indirectly by what they have in stock‟
(Andersson, et al., 2005, p. 342). This is in line with the results of Heikkilä, et al. (2010) They
found that the second most given reason to switch to generics is the advice of the
pharmacists. The savings of the introduction of the generic substitution were huge in
Sweden. Of all the substitutions, fewer than 5% was to a more expensive product. The
possible additional savings were substantial, but still not fully implemented. But not only the
introduction of generic substitutions played a role. Because of this introduction, brand name
manufacturers lowered their prices, to compete with generics. This is a by-effect of the
generic substitution in Sweden. In total, 60% of total savings was achieved. Andersson, et al.
further refers to a study undertaken in America (New Yersey). During generic substitution,
77% of the prescribing doctors had approved substitution and 97% of the patients who had
been offered substitution had given their consent (Suh, in Andersson, 2005).
Furthermore, they found that opinions in Denmark about generic substitution differ between
patients and doctors. And contrary to common wisdom, most patients are satisfied with
generic substitution, but two out of three doctors are not (Andersson, et al.). An explanation
was not given .
In 2007 Andersson, Bergström, Petzold & Carlsten looked at pharmaceutical sales, which
they divided into three categories; over the counter sales, hospital sales and outpatient
prescription sales. Their main conclusion was that when generic substitution was introduced,
the trend of increasing pharmaceutical expenditures was reversed into a decrease. However,
these results have to be interpreted with caution. The study period was 5 years, so it is too
early to conclude that this shift will last. One of the results of this decline is that the price per
unit could be reduced by promoting the use of cheaper generic equivalents through generic
prescribing, generic substitution and generic dispending.
The article of Mott and Cline (2001) examined the prevalence of prescriptions that offer the
opportunity for generic drug use. One of their main conclusions was that the role of the
pharmacist is very important to increase generic drug use. Another conclusion was that, in
order to enlarge the substitution in generics, the focus has to be on the prescriber.
They furthermore conclude that the costs for prescription drugs are one of the fastest
growing expenditures in the pharmaceutical sector. As well as Suh (1999) indicated, generics
can become about 70% cheaper than the same brand name drugs. The generic substitution
rate (% brand name prescription orders eligible for substitution that are substituted with
13
Master Thesis Stefan van der Goes
generic drug products) grew between 1987 and 1994 from 22% to 41%, and between 1991
and it grew from 33% to 43,2% (Mott & Cline, 2001). One can see that this is a decreasing
growth. This may have two causes; the possibilities for generics have become less, or the
substitution hasn‟t reached its full potential. The last cause can be affected by the brand
name pharmaceuticals. With more promotion on their products, the increasing growth of
generics could come to an end.
Mott and Cline further found that the following factors may be associated with the possibility
of generic substitution: prescriber and pharmacist characteristics, drug insurance coverage,
patient characteristics, and drug characteristics that may be associated with the opportunity
for generic drug use and substitution.
Mott and Cline too refer to the incentive that prescribers and pharmacists have by offering
generics, due to their lower prices. This can result in more satisfied patients (they save
money without losing quality of their drugs) and more profit for the pharmacists.
According to Mott and Cline there are two characteristics that are important to divide. One is
to look if the drug is to treat acute or chronic diseases, second is the rate of time (years) that
the generic is available.
Banahan III and Kolassa (1997) found that not only patients need good education about
generics, but the prescribers need good education too. Out of their research two groups of
prescribers were identified, those who were positive about generics and those who were
negative. Most of the prescribers who were negative, didn‟t know much about generics and
the requirements of the Food and Drug Administration regarding generics.
Gartner and Kreling (2000) examined consumer perceptions for generics of different medical
conditions and the relationship between risk perception and the cost savings. They conclude
that patients are sensitive for financial incentives when using generics. This even holds for
patients who find generics riskier than brand name drugs. Obviously, when there is a greater
(perceived) risk, the financial incentive has to be bigger to convince a patient.
A recent study done by Osinga (2011) claims that advertisements determines the choice for
a drugs. One of his results was that pharmacists usually stop to promote a drug when the
marketing activities of that drug are stopped (often when the patent is expired). General
practitioners then usually prescribe a (new) drug which is being promoted, instead of the
cheaper generic. When costs could be saved to prescribe the generic, this is usually not the
case. But, it must be said that the research by Osinga was conducted in the United States,
where different rules with respect to promotion of generic drugs are than in the Netherlands.
The next paragraph will discuss the Dutch rules and regulations concerning generic drugs.
14
Master Thesis Stefan van der Goes
2.1 Dutch policy on generic drugs
The Dutch authorities stimulate the use of generic drugs. This is due to, as mentioned
before, the increasing costs of health. This rise has two main reasons: more (complicated,
thus more expensive) drugs and the aging of the population. Not only will this aging continue
the coming years, it will accelerate too (CBS, 2011, p. 11). If there would be no stimulation of
generics, the costs of drugs would rise each year with at least 10% („Betaalbaar houden van‟,
2011). The Dutch government works together with insurance companies to reduce this costs.
The ministry of Health, Welfare and Sports is responsible for the policy about (generic)
drugs. Several measures have been taken to set a hold on the rising costs and try to get a
downward trend in these costs. One measure is to reimburse only the cheapest drugs,
containing the same working ingredient („Betaalbaar houden van‟, 2011). Health insurers are
free to choose if they want to reimburse more expensive drugs. This involves less costs for
health insurers and for their clients (sometimes depending on one‟s policy). Furthermore,
there‟s a list prepared by the cooperative health insurers (college van zorgverzekeraars)
about which drugs are being reimbursed by which health insurer. One aspect of this so called
preference policy is that the patient always has the last say about his medication. If a patient,
despite of what doctors or pharmacists might say, insists he only wants the brand name
drugs, pharmacists are obliged to offer this drug to him. For more information on the Dutch
policy to reduce costs regarding drugs, see Appendix D.
Furthermore, there is a site by run by the cooperative health insurers called
www.medicijnkosten.nl where one can compare the prices of different drugs containing the
same working ingredient.
Moreover, commercials for drugs are under restrictions of the government. The Inspection for
Public Promotion of Drugs controls laws made by Dutch and European governments.
First of all, only non-prescribed drugs are allowed for public promotion. Secondly, the drugs
can‟t make claims to cure, threat or prevent certain diseases. It is allowed to make claims
such as good for the hearing, good for a fresh breath or increases the resistance against
bacteria and fungi. There‟s a whole list of what‟s forbidden and allowed to claim, specified by
therapy area, which can be found at
http://www.koagkag.nl/content/index.php?option=com_wrapper&Itemid=216
This implies that manufactures of generic drug can only advertise for non-prescribed drugs.
This market is, due to competition, not a very attractive market. For painkillers for example,
different drugs are available from different suppliers, which implies that the margin would not
be very high in such a market. Marketing on generic manufacturers focuses mainly on the
prescribers of drugs, doctors and pharmacists. But like public promotions, there are rules for
these promotions. The rules are imposed by the Dutch and European government and
15
Master Thesis Stefan van der Goes
controlled by Foundation for Pharmaceutical Advertising. These rules are mainly about what
prescribers are allowed and not allowed to do or to how behave regarding promotions of
(generic) drugs.
These rules can be found on http://www.cgr.nl/267/Wet-en-Regelgeving.html
2.2 Generics and Private Labels
Because there is not so much research on (attitudes towards) generics, I decided to look at
research concerning private labels. Research on private labels (store brands – i.e.
Euroshopper, AH huismerk, etc.) and generics has shown that they are similar in various
ways. Both are „copies‟ of the original product, both are generally cheaper than the original
(up to 70% of the original price, Suh, 1999). And both their market shares are growing5.
Most research on private labels is concerned with private labels in retail. Generics are
(except some non-prescribed drugs) not available in retail stores. For buying generics, you
fully have to trust the drug, which is important, because one wants to cure from a disease or
to have less health issues. With private labels this is different, there is no real trust issue.
When someone buys a private label which turns out to be not as good as the original, one
can easily switch back (again) and buy the brand name product.
With generics this is different, because the market for pharmaceuticals isn‟t comparable with
the market for fast moving consumer goods (FMCG). Another difference is that generics can
be reimbursed (which and how much depends on one‟s health insurer) and this isn‟t the case
for private labels. But private labels can be bought for the whole family, or multiple persons in
a house, and generics are bought for one consumer only. Moreover, generics can be
prescribed (so no choice has to be made) and this is definitely not the case for private labels.
Furthermore, private labels can help to promote the store, but they are mostly used to create
or sustain customer loyalty.
Research conducted by Rubel (1995) seemed most relevant to my research. This study
shows that the private label brand sales in drugstore are highest in the categories: cold/flu
products, cough drops and syrups, and vitamins and headache remedies. Sometimes a trust
issue does occur with private labels. This is the case when there is, for instance, a sickness
of animals which can influence the food. The BSE-crisis („gekke koeienziekte‟) and the
dioxine in chickens are examples of this. To overcome this problem, private labels are
5 In 2002 in the USA, 20% of all grocery sales was private label products (Thompson, in Batra & Sinha, 2000), good for more than $48 billon
(Batra and Sinha, 2000). Forecasts are that private label might attract 40% or more of US supermarket sales (Denitto, in Batra & Sinha,
2000). In 2000, the sales of private label brands in grocery outlets in the USA exceeds $48 billion (Batra & Sinha, 2000)
16
Master Thesis Stefan van der Goes
sending out a message that they have to meet the same safety regulations as brand name
products do.
Research by Van Horen, Pieters and Stapel (2009), is concerned with copycats and the
consumer mindset. They concluded that copycats who look just like brand name products will
be less sold compared to products that have subtle imitations. This can be related to a
successful part of the brand-name product. This effect happens regardless of the mindset the
consumer has (judge mindset or consumer mindset). This is useful information for
manufactures of generics too. The common wisdom of a good marketing strategy for
copycats was to copy the A-Brand (or brand name drug) as much as possible. This research
has shown that that isn‟t the best solution. Or, to say it in other words, products do not only
need to have points of parity, but points of difference as well6.
Vaidyanathan and Aggarwal (2000) examined implications of ingredient branding for national
and private label brands. With ingredient branding they mean (in this case) that a private
label uses a national brand ingredient (that is widely known) in their product. In this research
they take the example of breakfast cereal (private label) with SunMaid raisins (national brand
ingredient). Another example of ingredient branding is found in the ICT business. Lots of
computer ads (on TV, radio, internet of in the store) are saying „Intel Inside‟. Thereby they
mentioning that they use a top A brand ingredient (the Intel chip) in their computers. This
works particularly for computers who have a brand name which is not known. Another
example is those computers run on Microsoft‟s Windows system. Batra and Sinha (2000) too
came up with this solution. They argue that third-party endorsement (i.e. ingredient branding)
will reduce consumers uncertainty about the quality of (in this case) private labels.
Vaidyanathan and Aggerwals (2000) main conclusion was that a private brand with a
national brand ingredient was evaluated more positively (or: the attitudes towards this
product are more positive than towards the same new private label product without the
national brand ingredient). This is due to the fact that when a new product is introduced by
an existing brand (i.e. brand extension), consumers tend to evaluate this new product by
using their existing value perceptions (as they relate to the original product) to evaluate the
new product.
A private label with a national brand ingredient is for both parties an effective way of
marketing. Private labels do not need to use lots of money to win people‟s trust and become
widely known. By the national brand ingredient people will evaluate the product more
6 Points of parity: make clear to consumers that your brand does well enough on a given attribute or benefit. Once this is established, it is
important to distinct from the other product. This is done with points-of-difference. With points-of-difference a brand can demonstrate
clear superiority. (Kotler and Keller, 2009)
17
Master Thesis Stefan van der Goes
positively and thereby releasing it from one of the private labels images: low quality. The
national brand ingredient will act as a quality signal.
National brand producers have advantages of such collaboration as well. They will get more
sales, they will sell to bigger markets, they are better off when purchasing ingredients
(because of economies of scale) and they will have wider promotion. The authors state that
there is no danger of cannibalization, but that this ingredient branding leads to additional
sales. They think that the potential consumers of the new product will be consumers who
already buy private labels. They will not target at national brand consumers (due to price).
Still there is nothing mentioned about loyalty buyers (consumers who buy a product just
because of its name). They might shift to this new product, because the name of a national
brand they used to buy is in it.
Their research further showed that the image of the national brand wasn‟t hurt by the
cooperation with the private label. It seems that the national brand joining with the private
label actually helped perceptions of the national brand by value conscious customers.
Although the conclusions of this research are interesting, it is important to know that this
research is done under the assumption that the price of the private label product did not
raise when adding a national brand ingredient.
Burton, Lichtenstein and Netemeyer (1998) looked for their research to a scale for measuring
attitude towards private label products. Or, to say in other words, they examined which
factors determine someone‟s attitude towards private labels. The authors focused on the
attitude of the consumer towards private labels product as a whole. They conducted their
research towards private labels in the retail business. Much of the literature suggest that
attitudes towards private labels can be roughly divided into three categories: consumer price
perceptions, marketing constructs and deal proneness constructs. The figure on the next
page may clarify this model.
18
Master Thesis Stefan van der Goes
Figure 1: schematic model of influences of attitudes towards private label products (Burton, et al., 1998)
One of the conclusions was that attitude towards private labels was positively related to deal
proneness, value consciousness7, reliance on internal reference prices and smart-shopper
self-perceptions, but negatively to brand loyalty, price-quality perceptions and impulsiveness.
Furthermore, they found that risk averseness was not related to private label attitude. But
brand loyalty, impulsiveness and smart shopper self-perceptions were of influence towards
consumers attitudes towards private brands. Another conclusion was that private label
attitude is positively related to the percentage of private label purchases made on a shopping
trip and can explain price perceptions, deal perceptions and other constructs. This last
conclusion was by far the weakest, because this was measured on only one shopping trip.
The authors further found that, by looking at the behavioral level, a negative association can
be found between private label purchases and purchases using price discounts or coupon
purchases.
Garretson, Fisher and Burton (2002) wanted to know why price oriented consumers have
different attitudes towards private labels and national brand promotions. One finding was that
value consciousness is positively related to attitudes towards private labels and national
brand promotions. But a negative effect was found too. The attitudes of consumers who see
price as a indication of quality, were negative towards private labels. The national brands on
promotion were seen as a way to achieve savings, there was no feeling of loss of quality.
Another finding is that the consumers who are value-conscious may be favorably
predisposed both to national brand promotions, private labels or a combination of both.
7 As defined by Lichtenstein, Ridgway and Netemayer (in Burton, et al., 1998): ‘a concern for paying low prices subject to some quality
constraint’
19
Master Thesis Stefan van der Goes
Third, they found that smart-shopper self-perception is of influence on attitudes towards both
private labels and national brands. Smart-shoppers are obviously interested in saving
money, but even more how they can save money. For instance, the relationship of smart
shopper self-perception and national brand promotion is significantly more positive than the
relationship between smart shopper self-perception and attitude towards private labels.
The consumer who is looking for pure value, or have no pretentions towards their shopping
acuteness, finds that private label brands satisfies their needs.
Garretson, et al. (2002) as well looked at how the quality perceptions of private label brands
can be improved.
One important conclusion is that upgrading the tangible quality of the product (packaging)
can be profitable. One other solution is to continue improve the product. Another conclusion
is that it is important to educate consumers about „how the quality is built into the product‟.
Batra and Sinha (2000) presented this solution too. They say that (in order to sell more
private label brands) retailers need to put as much objective information about the product
ingredients and manufacturing quality as possible on the package label; thereby reducing the
uncertainty that the consumer might have.
Garretson, et al. (2002) conclude that attitudes towards private labels and national brand
promotions are, not only influenced by value-consciousness but price-quality associations,
smart shopper self-perception are of influence too.
Glémet and Mira (1993) found that categories with high private label brands were those that
provided, among other characteristics, an easy comparison.
Batra and Sinha (2000) found that as the consequences of making a purchasing mistake
decline, the demand of private labels increases. The perceived consequences of making a
purchase mistake are higher when the different brands in the category are seen as differing
considerably in quality. The probability of making a mistake can be reduced by educating the
consumer, which can be done by i.e. ads or packaging.
2.3 Attitudes and trust towards generics
What determines one‟s attitudes towards generics can be dependent of multiple variables.
One of these factors is trust. When people don‟t trust generics, obviously, their attitudes
would be more negative compared to people who do trust generics. So, trust could be
viewed as an antecedent of attitudes.
Furthermore, people sometimes might have difficulties to describe their attitudes, as
„attitudes‟ might sounds too vague. When referring to trust is would be easier for the most
people to understand. For these reasons, the respondents were asked how much they trust
generic drugs.
20
Master Thesis Stefan van der Goes
Finally, the following table provides an overview of the main conclusions of the discussed
literature.
Author(s) Main conclusions
Kjoenniksen, et al. (2006) - 2/3 of people who had their drug replaced by generics claimed overall satisfaction
Suh (1999) - high growth in purchase of generics - in three years the price of a generic drops to 25% of the brand
name equivalent - generic substitution can be maximal when general practitioners,
pharmacists and insurance companies are involved Andersson, et al. (2005) - pharmacy personnel is a important factor in generic substitution
- large savings in public health in Sweden due to generic substitution - because of the generic substitution, brand name manufacturers
lowered their prices - in Denmark, most patients are in favor of generics, but two out of
three doctors are against Andersson, et al. (2007) - with the introduction of generic substitution, the increasing
pharmaceutical expenditures was reversed into a decrease Mott and Cline (2001) - the role of the pharmacist is very important to increase the use of
generics - in order to enlarge generics, the focus has to be on the prescriber - the costs for prescription drugs are one of the fastest growing
expenditures in the pharmaceutical sector - important characteristics are acute or chronic diseases and rate of
time (years) that the generic is available Banahan III and Kolassa (1997) - not only patients but prescribers too need good education of
generics Gartner and Kreling (2000) - patients are sensitive for financial incentives when using generics Osinga (2011) - advertisements determines the choice for a drug
- general practitioners usually prescribe a (new) drug which is promoted, instead of the cheaper generic
Table 2.3.1: Overview of literature
21
Master Thesis Stefan van der Goes
2.4 Hypotheses
When looking at the questionnaire (see Appendix B) the questions can be divided between
demographics and (shopping) behavior. At the end of this study it will be clear which
demographics are of influence on attitude towards generics and which (shopping) behavior is
of influence on attitude towards generics. Figure 3 illustrates this combined focus.
This chapter will describe the different hypotheses that will be tested. Furthermore, the
background and relation to attitudes towards generics is being discussed.
2.4.1 Gender
When developing a new drug, pharmaceutical companies usually test a new drug only on
men (Moyer, 2010). This is quite remarkable since there are obvious differences between
men and women (e.g. in hormones, genes, brains, etc.). The reason for testing on men only
is that men don‟t have a menstrual cycle and therefore their hormones don‟t fluctuate. So,
men are a more homogenous group than women and the results of the tests are easier to
interpret. A result is, however, that the differences between men and women aren‟t taken into
account8. For instance, women respond different on drugs like antidepressants and
antipsychotics (higher concentration of the drug in their blood). For influenza vaccine, women
could do with half of the doses.
Men and woman furthermore differ with respect to the frequency of drug usage. For instance,
woman use 2.4 times more sleeping pills than men. Men use 1.2 times more cholesterol
lowering drugs than women (Stichting Farmaceutische Kengetallen, 2005).
8 A well-known example is what happened in the 1960s. The drug Softonol (a tranquilizer) was only tested on men, but side
effects were that pregnant women who had used the drug gave birth to misshapen children (http://www.kennislink.nl/publicaties/softenon-vloek-en-zegen, retrieved on 13-07-2011). Nowadays, Softonol is used to treat leprosy (Boguski, Mandl and Sukhatme, 2009)
Figure 2: schematic research model
22
Master Thesis Stefan van der Goes
Furthermore, women use 54% more drug than men do9. This could mean that women face
information about generics more often, which might influence their attitudes towards
generics.
A study by Loyd and Gressard (1984) on attitudes towards computers showed no gender
effect. But their study focused only on (high school and college) students.
Negative experiences of the switch from brand name drugs to generics were not related to
gender, according to Kjoenniksen, et al. (2006). A previous study by Burton, et al. (1998)
found a relationship between gender and attitudes towards private labels.
Ailawadi, Neslin & Gedenk (2001) detected that women are more likely than men to be
innovative, impulsive, shopping mavens, planners and more store loyal. Store loyalty could
mean that if a pharmacist advises a generic drug, they will accept is, because they are loyal
to the store (pharmacy).
There is mixed evidence with respect to gender and attitudes towards generics. Therefore it
is unclear whether a difference between men and women on attitude towards generics exists
and what this difference might be. Therefore, hypothesis 1 is formulated as a two sided test:
H1: The attitudes towards generics are dependent of gender
2.4.2 Age
As people get older, they have more health-issues, and will therefore use more drugs. Data
from the National Institute for Public Health and Environment (RIVM) showed a big difference
between the group of 65 years and older and other age groups with respect to prescribed
drugs10. The group of 65 years and older get more drugs prescribed compared to the other
groups (at least 10% higher than the second highest age group).
For brand name and generic manufacturers it is interesting to know whether there are
differences among different age groups and their attitudes towards generics. This helps then
to determine on which age group they should target their marketing/research activities.
Moreover, differences between age groups and attitudes towards generics could be caused
by the use of more drugs when people are getting older.
Richardson, Jain and Dick (1996) argue that the propensity of consumers to buy private
labels depends on various demographic factors and one of them is age.
Age is a factor that determines attitudes towards generics, according to Kjoenniksen, et al.
(2006). Patients in Norway of 50 years and younger were 3,7 times more likely to change
9 http://www.joop.nl/leven/detail/artikel/vrouw_gebruikt_54_meer_medicatie_dan_man/ retrieved on 09-07-2011
10 http://www.rivm.nl/vtv/object_document/o4982n25485.html
23
Master Thesis Stefan van der Goes
from a brand name drug to a generic drug than those who were 70 years and older. Another
finding of Kjoennisken, et al. (2006), was that negative experiences from a switch to generics
were not related to age.
Mott and Cline (2002) found that the „inflection age‟ (possibility for generic substitution) is 57.
Contrary to these findings are results of a study by Burton, et al. (1998). They concluded that
differences for age in relationship with attitudes towards private labels were not significant.
As these several studies concluded that older people were more averse against generics, the
following hypothesis was formulated:
H2: Age has a negative effect on attitudes towards generics
2.4.3 Type of environment
It is a worldwide trend that more and more people migrate from a rural environment to an
urban environment (Fields, 1975). The Netherlands are no exception on this trend. in 2005,
more people lived in an urban environment than in a rural environment for the first time in the
Netherlands. (“Meer mensen in”, 2006). An urban environment is largely created through
urbanization. Several studies have examined the difference between people living in a urban
environment, a semi-rural environment and a rural environment.
Urban areas have more health facilities, and more specialists, so knowledge about generics
is higher in those areas. The definition „urban‟ or „rural‟ can be of great importance of an area
because it has lot of (policy) implications. For instance, a government could determine that a
certain urban area there should have a certain minimum number of hospitals, ambulances,
pharmacies, etc.
The CBS defines rural as: “an area with less than 1000 addresses per square kilometer”
(http://www.cbs.nl/nl-NL/menu/methoden/toelichtingen/alfabet/p/platteland1.htm). When an
area has more than 1500 addresses per square kilometer, it is defined as urban. When an
area has 1000-1500 addresses per square kilometer it is defined as semi-rural.
Figure 2 provides an overview of the population density by town in the Netherlands.
24
Master Thesis Stefan van der Goes
A study of Paykel, et al. (2000) on
urban-rural mental health differences
in Great Britain showed an
association between mental health
conditions and the type of area
(urban/rural). The results of this
study included that: The rates of
psychiatric morbidity, alcohol
dependence and drug dependence
were higher in urban settings than in
rural settings. The semi-rural setting
was intermediate. Other differences
between urban and rural settings
were that the population in urban
areas was significantly younger, not
currently married, of lower social
class, non-white, less well-educated,
living in flats of non-detached houses, a lower proportion of which were owned outright.
Again, semi-rural settings were intermediate.
Citizens of urban settings are for instance more likely to have experienced a stressful life
event in the last year. This stressful event can cause these people to have a (prescribed of
non-prescribed) drug.
For alcohol and drug dependence and one‟s area, no effect was found.
A study by Peen, Schroevers, Beekman and Dekker (2009) showed that the amount of
people having one or more physical problems was 80% higher an urban environment than in
a rural environment.
People in urban areas are having more diseases and their use of drugs will be higher. This
can lead to more usage and/or knowledge of generics. As the lower income of urban areas
can be an incentive to use generics, I expect that people in urban areas have a more positive
attitude with respect to generics. The following hypothesis will be tested:
H3: People in urban regions will have a more positive attitudes towards
generics than people living in rural areas
Figure 3: Population density (number of inhabitants per square kilometer) in 1999. The darker the blue, the more inhabitants a town has
Source: http://www.compendiumvoordeleefomgeving.nl/indicatoren/nl2101-
Bevolkingsdichtheid-Nederland.html?i=15-12.
25
Master Thesis Stefan van der Goes
2.4.4 Knowledge about generics
Every drug one has, comes with a package leaflet. The package leaflet contains general
information about the drug, the possible side effects, how much doses one must take, etc.
Via this package leaflet people will have more information about the drug and might therefore
will use it wiser (besides, this information can lead to more knowledge about the product). It‟s
Interesting to examine whether people who had been given information of generics (via a
package leaflet or other ways) have a more positive attitudes about generics. An article of
Kjoenniksen, et al. (2006) showed that patients who had received information about generics
were more likely to have switched. About 2/3 of the patients claimed overall satisfaction,
about 1/3 of the patients who had switched reported a negative experience due to the
substitution. These results were independent of age, gender, number of medications or the
physician. Kjoenniksen, et al. concluded that additional information and support from
physicians and pharmacists is needed to help the acceptance rate of generics grow.
Furthermore, there was a clear statistical correlation between generic exchange and whether
the patient had been given information from their physician (Kjoenikksen, et al.). The same
relationship was found between information given at the pharmacy. According to
Kjoenniksen, et al. the highest substitution ratio was provided when a patient gets
information given by the general practitioner and the pharmacy.
Kjoenniksen, et al. stated that more information will positively influence the accepting rate of
generics. Whether such relation(ship) does exists in the Netherlands will be tested with the
following hypothesis:
H4a: Knowledge about generics has a positive effect
on attitudes towards generics
2.4.5 Use of generics
Although generics are relatively new in the Netherlands, people are already using them. Most
health insurers already only reimburse the cheapest medication, thereby supporting the use
of generics. Use of generics can imply that people become familiar with them, which might
influence their attitudes towards generics. Caspi (1984) examined the attitudes of children
towards the elderly. His main conclusion was that children who had daily contact with elderly
persons held very positive attitudes towards them, whereas children without such contact
held vague or indifferent attitudes.
Batra and Sinha (2000) have found an reverse effect on private labels. In describing factors
leading to the buying of private labels, experience characteristics lead to higher perceived
quality variation and higher felt consequences of making a purchase mistake. These factors
26
Master Thesis Stefan van der Goes
reduced the purchasing of private labels. Research by Kjoenikksen, et al., (2006) suggested
that 2/3 of patients who had switched to generic drugs were satisfied with these drugs. To
test whether use of generics has an positive influence on attitude towards generics,
hypothesis 4b will be tested:
H4b: Currently using generics has a positive effect
on attitudes towards generics
2.4.6 Social Economic Status
H3 was concerned with the question whether the type of area had an influence on attitudes
towards generics. Next to someone‟s living area, Social Economic Status (SES) can
influence attitudes towards generics too. It is not entirely clear of which items SES should be
composed. There are multiple definitions of SES containing different items. However, income
and education are the most used SES-items.
Winkleby, Jatulis, Frank, E. and Fortmann (1992) concluded that the best predictor of a good
health is higher education. Income, occupation and risk factors to cardiovascular diseases,
were not significant. According to Adler, et al. (1994) a relationship between someone‟s
health behavior and level of education exists. Men and women with a lower education do
smoke more, have less physical activity (which, for example, increases the likelihood of
obesity). A different relationship was found for the use of alcohol. A positive correlation with
SES (measured by job status and education) and alcohol was namely found. One must
however be careful with drawing conclusions. Alcohol can do harm (i.e. liver problems), and
do good (i.e. moderate levels of alcohol decrease the chance for coronary heart diseases).
Furthermore, Adler, et al. (1994) concluded that people of the lower classes of SES had the
highest rates of morbidity and mortality of all SES classes.
Figures from the CBS (2011, p. 14) confirm this. Higher educated people usually maintain
healthier lifestyles than lower educated people. This leads to their conclusion that at 65
years, higher educated people have generally 8 years more to came up with a good health
than lower educated people.
A rapport of the National Institute for Public Health and Environment (RIVM) in 2002 showed
a relationship between SES and accessibility to healthcare. Some of the results included:
people of lower SES had more contact with their general practitioner, they use less non-
prescribed drugs and participate less in screeningprograms. For contact with physiotherapist,
visits to the dentist no relation(ship?) was found. Their SES contained income, education and
occupation.
27
Master Thesis Stefan van der Goes
The National Public Health Compass (Nationaal Kompas Volksgezondheid), part of the
ministry of Public Health, Welfare and Environment indicated the existence of a relationship
between SES and health too. Lower-educated men and women live (on average)
respectively 4,9 and 2,6 years shorter than men and women with a high education. On
average, lower-educated people live 15 years more in poorer health (RIVM, 2006).
Furthermore, the National Public Health Compass found that lower educated people get
more drugs prescribed than higher educated people, even when controlling for age, gender
and self reported health (RIVM, 2010).
As people from lower SES have more health issues, they will probably use more drugs. So,
the chance that they are confronted with generics will be higher than in other groups. People
with a lower SES, had lower incomes too. This will lead to more price sensitivity, making
generics more interesting. Therefore, I expect that this would positively influence their
attitudes to generics. This is tested by means of the following hypothesis:
H4c: People with a lower SES will have more positive attitudes towards
generics than people with a higher SES.
2.4.7 Type of disease
The above hypotheses are all about demographics and how these could influence attitudes
towards generics. Having a chronic disease could be a characteristic which could have an
effect on attitudes towards generics as well. People who have a chronic disease might
possibly use more drugs (hence, they could have been given information by for example the
pharmacists or health insurer) and will use these drugs for a longer term. The large drugs
usage could lead to familiarity with a drug and aversion to switch to the generic equivalent.
Chronic diseases are less likely to be generically substituted. Possibly because consumers
would less willing want to use generic drugs for treating chronic conditions (Mott and Cline,
2002).
People using many different drugs were likely to have more (chronic) diseases compared to
people using fewer drugs, according to Kjoenniksen, et al. (2006). They furthermore
concluded that people using many different drugs are 2.6 times more likely to have switched
to generic drugs. They further found that negative experiences due to the switch from brand
name drugs to generics were not related to polypharmacy (people using many different
drugs).
So, having a chronic disease could influence attitudes towards generic drugs.
28
Master Thesis Stefan van der Goes
Most studies indicate a negative relationship between having a chronic disease and attitudes
towards generics. Whether this holds for the Netherlands too is tested via the following
hypothesis:
H5: People having a chronic disease will have a more negative attitude
towards generics than those who don’t
2.4.8 Brand Loyalty
People can have strong feelings for a particular brand and will prefer this brand above other,
comparable brands. This effect is largely researched in fields like retail, but brand loyalty can
occur when people buy drugs too. Garretson, et al. (2002) describes (the behavior of)
consumers who are loyal to brands as follows: „Loyal consumers are more likely to pay full
price for their favorite brands and look for them in any store they shop. If not found, they may
shop elsewhere‟ (Garretson, et al., 2002, p. 92).
Their general conclusions were that consumers tend to be less loyal to products in markets
were many brands are available, where number of purchases and dollar expenditure per
buyer are high, where prices are relatively active, and where consumers might be expected
to simultaneously use a number of brands for the product. According to the same study,
consumers actually are brand loyal in markets where brands tend to be widely distributed
and where market share is concentrated heavily in the leading brand.
However, research conducted by Gartner and Kreling (2000) suggested that patients are
sensitive for financial incentives when using generics. This even holds for patients who think
that generics are riskier than brand name drugs. Obviously, when this is the case, the
financial incentive has to be bigger. This means that brand loyalty for brand name drugs only
holds when price incentives are not high. Buying drugs thus can be due to brand loyalty, but
can occur of a result of inertia as well.
Brand inertia can also be defined as short term brand loyalty (Jeuland, 1979).
Contrary to common wisdom, brand loyal tendencies people not to have negative attitudes
toward promotions of private labels (Garretson, et al. 2002). Furthermore, Garretson, et al.
(2002) revealed that brand loyal consumers were only interested in price savings for „their‟
brands. These consumers use these price savings to stockpile their inventory.
29
Master Thesis Stefan van der Goes
One could state that when one uses a particular drug for a long time, one could become loyal
to that brand. Especially with medication, people don‟t want to take any risks and they prefer
to rely on brands they trust. If the previous studies are right, brand loyalty will have a
negative effect on attitudes towards generics. The following hypothesis was formulated:
H6a: Brand loyalty has a negative effect on attitudes towards generics
2.4.9 Price, (perceived) Quality and Status
Brand loyalty is not the only factor related to (shopping) behavior which could influence one‟s
(shopping) behavior. Price, (perceived) quality and status are other shopping related factors
which could have an effect on attitudes towards generics. This paragraph will discuss those
three characteristics.
Szymanski and Busch (1987) discussed the characteristics of the generic-prone consumer. A
strong relationship between perceived product quality, price and purchasing of generic
products was found. The authors argue that this means that the buyer of generics is
convinced that buying generic provides him good value for his money. More specifically, the
low(er) price is seen as the most significant benefit of generics. It should be noted however,
that the study by Szymanski and Busch was conducted in 1987, Back in those days the
supply of generics was much lower.
According to Cunningham, Hardy and Imperia (1982), 80% of the buyers of generics valued
price as the most important reason to buy. Heikkilä, et al. (2010) claims that the number one
reason to switch to generics is price.
Grossman and Shapiro (1988) found that consumers are willing to pay more for counterfeit
products than for generic merchandise of the same quality. This is due to the fact that
consumers are willing to pay extra for the prestige (status) associated with brand name
trademarks. One of the results of a study by Hoch and Banerji (1993) was that consumers
bought private labels merely because of their low prices. Quality was the most important
reason to buy (or not buy) a private label. When consumers were aware that the private label
offers the same quality as the national brand, the price factor played a role and for this
reason some consumers favored private labels.
Rosen (1984) did found similar results. Respondents had to compare overall quality, quality
consistency over repeat purchases and quality similarity across stores. In all three areas,
generics were perceived to have the least quality. Private labels are often seen as „inferior‟ to
the A-brand.
30
Master Thesis Stefan van der Goes
As these studies showed, a relationship exists between price, quality, status and attitudes
towards generics. Whether these factors influence the attitudes towards generics in the
Netherlands too is tested with the following hypotheses:
H6b: A lower price will lead to a positive effect on attitudes towards generics
H6c: (perceived) Quality has a negative effect on attitudes towards generics
H6d: Higher status has a negative effect towards generics
Some of the hypotheses mentioned in this chapter predicted a positive relationship, same a
negative relationship and some hypotheses did not include a prediction regarding the nature
of the relationship. The figure below provides an overview of the research model.
Figure 4: schematic overview research model
31
Master Thesis Stefan van der Goes
3. Methodology
To obtain answers to the different hypothesizes, a survey was conducted. People were
asked to fill out an online questionnaire. It was anticipated that the respondents not all have
enough specified knowledge about the difference between generic and branded drugs, some
extra information was provided in the above the first question about generics. questions
about generics will be better understood when there is a text above the question which one
can read at his own speed (and over and over again if it is needed), instead of providing this
information in an interview. The risk of an interview is that people will say they understand
the question when they actually don‟t understand, just because they are afraid of looking
stupid. This behavior will obviously lead to answers that are not reliable.
In the questionnaire, control questions were asked (on a reverse scale)11 to check for
reliability.
Before the questionnaires were sent, a pre-test of the questionnaire was conducted. The
results of the pre-test were that questions 22, 25 and 26) needed more information.
Secondly, some more questions were added to get more information about people having
certain demographics. The pre-test was always completed in no more than 15 minutes. 10
People (relatives) were included with the pre-test.
After the improvements had been made, the questionnaire was sent by mail to friends,
family, colleagues and bachelor I students of the ESE who followed the course Marketing
and for which I worked as a student-teacher. A reminder was sent to make sure that
everyone filled out the questionnaire. Furthermore, to get enough respondents in every
group, the questionnaire was sent to a randomly selected group of 50 people of 45 years of
older.
The final questionnaire consisted of 34 questions about demographics and behavior. For
most of the demographics (age, gender, residence, type of disease, income), a respondent
could click a box to indicate in which class he or she belonged. For measuring the different
categories, the classification of the Central Bureau of Statistics was used.
The questions regarding behavior were asked on a 5-point Likert scale. Respondents could
indicate how their preferences were on various questions and how much they (dis)agreed
with a given statement.
11
The control questions were I am sensitive to sales (Q29) and I love to try new brands (Q29).
32
Master Thesis Stefan van der Goes
There was a chance that not all respondents were familiar with generic drugs. To overcome
this, the first 6 questions were about private labels in supermarkets. After that, a text followed
which contained information about generics. Especially the link between private labels and
generics could provide some practical examples for those who were unfamiliar with generics.
Just to make sure that nobody was unfamiliar with private labels, there was a text provided
after question 1 which contained information and examples of private labels. These
examples came from the largest supermarket chain in the Netherlands, Albert Heijn, but
from another supermarket too, C1000.
To make sure all the respondents knew the difference between prescribed and non-
prescribed drugs (questions 25 and 26), information and examples where given. Aspirin was
used as an example, as this is maybe the most widely-known non-prescribed drug.
Furthermore, there was an explanation at question 25 of the differences between prescribed
and non-prescribed drugs. What was meant by chronic disease was explained at question
19.
The answers to the questions about private labels were not used in the factor analysis. The
reason for this is that attitudes towards private labels could differ from generics. This could
blur the results of the factor analysis.
For hypotheses 1, 4a and 4b an independent samples t-test was used. Hypothesizes 2, 3
and 5 were tested with an Anova test. For hypothesis 6a, 6b, 6c and 6d a factor analysis was
done, followed by a linear multiple regression.
33
Master Thesis Stefan van der Goes
4. Results
4.1 Data exploration
Before the analysis started, the data was explored. There were 205 respondents, of which
132 did completed the questionnaire. 39% (n=69) Did not complete the questionnaire. The
majority of these respondents were not excluded from the dataset, because the answers they
did give were relevant for certain hypotheses. Replacing the answers of the 69 non-
completers with the mean was no option, because that would bias the results too much.
Of this group of 69 respondents, three respondents didn‟t fill in the questionnaire honestly
(i.e. „funny‟ answers, or at every question the same answer). Those 3 cases were excluded
from the dataset.
Secondly, the exploration of the demographics took place. Most of the respondents live in
rural regions (n=71), followed by urban regions (n=42) and semi-rural regions (n=18). In the
questionnaire, respondents were asked to type in their zip-code, because not everybody will
know if he or she lives in a rural, semi-rural or urban place. The zip-codes were computed
into dummies for the three different regions. For this computing, data and definitions of the
CBS were used. Via the zip-code it was checked in which city one lived. For this city, the
population (numbers of 2010) and the square kilometers were looked up. Lastly last the
population was divided by the square kilometers and the outcome was compared to the
definitions of the CBS for population density. For example: Westland (99.717 inhabitants /
79,52 km²) = 1.253,986 inhabitants per km² = semi-rural. A list of all the calculations can be
found in appendix C.
There were more females than males in my sample (91 vs. 41), and most of the respondents
(29%) had HBO as finished or current education. Furthermore, most of the respondents
(n=47) were between 45 and 65 years old. The group of 0-15 years consisted of zero
respondents and was left out of the analysis, the group of 65 years or older consisted of only
8 respondents. The choice was made to create a new group which consisted out of the
variables „45 to 65 years‟ and „65 and older‟, called „45 years and older‟.
Appendix A provides an overview of all the demographics.
Third, the data was explored to see if the assumptions for paramemetric data were met. To
check this, the Kolmogorov-Smirnoff test (Field, 2005) was used. For Education, Age and
Income the test was highly significant, which meant that the data are non-normal. Even after
transforming the data (log, square root) this result stayed. This is cause for concern, but due
34
Master Thesis Stefan van der Goes
to the large sample (n=136) this might not actually be a problem. But to be sure, non-
parametric tests were conducted when needed.
4.2 Gender
To test whether there was a difference in attitude towards generics between males and
females, an independent t-test was conducted. The non-parametric test (Mann-Whitney test),
was conducted too, because the data were not normally distributed and the groups sizes
differ. Theoretically the group sizes should be equal.
The result of the test were that males (N=41, M=3,51 SE=0,898) on average did not showed
a significant difference with females (N=91, M=3,40, SE=1,063) on attitudes towards
generics.
The Mann-Whitney test came to the same conclusion: on average, males (Mdn = 3) didn‟t
significantly differ with females (Mdn = 3) with respect to attitudes towards generics,
U=1790,000. Hence, hypothesis 1 is rejected.
4.3 Age
To test hypothesis 2, an Anova test was conducted. Ideally, the different groups have the
same sizes. As these are categorical variables, transforming the data by taking the square
root or log has no use. The results will be the same as for the non-transformed data. The
differences in group size could be significant. It‟s important if the variances of all groups are
the same. To test this, Levene‟s test was done to test of the variances of this three groups
differ significantly.
N Mean SE
15 – 25 years 32 3,50 ,718
25 – 45 years 44 3,66 ,914
45 years and older 52 3,42 ,997
Levene‟s test was not significant (p=0,286), so the variances of the groups are not
significantly different. So, the assumptions for the Anova test was not broken.
Looking at the means in table 4.3.1, it‟s clear that they are close to each other, so there will
probably no difference . To test whether there is a significant difference between the means
Table 4.3.1: Results of the Anova test
35
Master Thesis Stefan van der Goes
of the groups, post-hoc tests were used. The choice was made to use the tests of Hochberg,
Gabriel, Games-Howell and Dunnett. The reason for this was that Hochberg and Gabriel are
tests for when sample sizes differ. Hochberg is more accurate when sample sizes do differ a
lot, Gabriel is more accurate when sample sizes do slightly differ. It is recommended by Field
(2005) to use the Games-Howell test too, because of the uncertainty of knowing whether the
population variances are equivalent (Field, 2005, p. 341). And finally, Dunnett’s test was
chosen, because that‟s the only way to test the means of the groups against a control mean.
In this case, the mean of above 45 years had been chosen as control mean. All results of
Anova with post hoc tests can be found in Appendix F.
Post hoc test showed no significant results, so the conclusion was that the group of 45 years
and older did not had significantly more negative attitudes towards generics compared to
other groups, F(2, 125) = 0,823. Hence, hypothesis 3 is rejected.
4.4 Type of environment
Hypothesis 3 stated that people who live in an urban environment would have more positive
attitudes towards generic drugs than people living in a semi-rural or rural environment. To
test this hypothesis, like the test of hypothesis 2, an Avona with post hoc tests was
conducted. The post hoc tests were Hochberg, Gabriel, Games-Howell and Dunnett. „Urban‟
was used as the control group for the test of Dunnett.
Levene‟s test was not significant (p=0,182), there were no small groups, so no assumptions
of Anova were harmed.
N Mean SE
urban 42 3,17 ,148
semi-rural 18 3,78 ,191
rural 67 3,60 ,106
The Anova test showed a p-value of 0,018, which implies that not all groups have the same
means. To explore were the difference was, post hoc tests were conducted.
All the post hoc tests showed a significance difference between the means of the urban
group and the means of semi-rural and rural.
The results of the tests are quite remarkable. The hypothesis was formulated that people in a
urban environment would be more positive about generic drug, while in fact they‟re less
Table 4.4.1: Results of the Anova test
36
Master Thesis Stefan van der Goes
positive about generic drugs than the other groups! Appendix G showed that people in an
urban environment in percentage have more chronic diseases, which implies more use of
drugs.
To conclude: People in urban regions had on average a more negative attitude towards
generics than people in semi-rural and rural areas, F(2, 124) = 4,15, p < 0,05. As the
hypothesis was formulated that people in urban areas would had a more positive attitudes,
the hypothesis was rejected.
4.5 Knowledge about generics
An independent t-test was used to test hypothesis 4a. Because the groups are not equal the
Mann-Whitney test was conducted too. The results were as followed:
People who are unfamiliar with generics (N=88, M=3,50, SE=0,105) did not significantly differ
from people who are familiar with generics (N=44, M=3,50, SE=1,069) on their attitudes
towards generics, t(130) = 0,276, ns.
Just by looking at both means, which are almost equal, one could conclude that there will be
no difference. The p-value of this test is 0,276 means that the hypotheses of different means
must be rejected.
So, on average there was no significant difference between people who are familiar with
generic drugs (M=3,50, SE=0,983) and people who aren‟t familiar with generic drugs
(M=3,30, SE=1,069) and their attitudes towards generics, t(130) = 0,276, ns. Hence,
hypothesis 4a must be rejected.
4.6 Use of generics
To test hypothesis 4b, it is important to know who use(d) generic drugs and what their
attitudes towards generic drugs are compared to people who haven‟t used generic drugs. In
the questionnaire these groups were split in two. First, there was asked if a person is using
or in the past had used generic drug. If so, their attitudes towards generics were asked. A
problem arise when people had typed in that they were using or used generic drugs and filled
in that they never had used generics. This problem occurred 31 times. To see to which group
a person belongs, I looked at their answer to questions 12 of 13. Question 12 asked what the
reasons were to the use of generics and question 13 „if you never used generic drugs, what
could be reasons to use it in the future?‟, so when to had typed in only question 12 or 13 is
was clear to which group someone belonged. Still, there were 18 respondents who typed in
that they had and don‟t had experience with generic drugs and both filled in question 12 and
13. One way to solve this problem could be to look at question 8: „have you ever got any
drugs from another brand than you used to?‟. If the answer was no, this could mean that they
37
Master Thesis Stefan van der Goes
haven‟t got any generic drugs. But this isn‟t secure enough, those respondents could as well
been given generic all the time (if, for instance they use a kind of drugs frequently, but only
since two years – this could be a generic drug). So, for reliability reasons, this 18
respondents were not involved in the tests for hypothesis 4b.
When comparing the means of both groups, hypothesis 4b will be tested. An independent t-
test therefore was conducted. The results were that on average there‟s no significant
difference between people who use(d) generic drugs (N=51, M=3,57, SE=0,985) and people
who aren‟t familiar with generic drugs (N=63, M=3,46, SE=0,981) and their attitudes towards
generics, t(12) = 0,585, ns.
Because of unequal groups sizes, the Mann-Whitney test was conducted too. The Mann-
Whitney test resulted in the same conclusion as the independent t-test. On average, people
who use(d) generics (Mdn = 3) didn‟t significantly differ with people who never used generics
(Mdn = 3) with respect to attitudes towards generics, U=1493,000, ns. Hence, hypothesis 4b
must be rejected.
4.7 Social Economic Status
Hypothesis 4c tests the differences between multiple levels of Social Economic Status (SES)
and their attitudes towards generics. To test this hypothesis a new variable had to be
conducted. Taking the scores of „education‟ and „income‟, adding those up and dividing those
new scores by 2 gave the new variable. To make the different classes, there was chosen to
make four classes, given by the first, second, third and fourth quartile12. Due to this format,
no problems with Levene‟s test arose. The descriptives of the Anova test were the following:
N Mean SE
very low 20 3,50 ,224
low 28 3,29 ,169
high 23 3,74 ,180
very high 36 3,58 ,161
12
As there is no clear definition of about what classes SES should contain in the literature. And due to the fact that there are several small
groups in the variables ‘Income’ and ‘Education’, are used for this variable.
Table 4.7.1: Results of the Anova test
38
Master Thesis Stefan van der Goes
To test the hypothesis, Anova with post hoc tests was used. The Anova test was not
significant (p=0,367). Logically, none of the post hoc tests (Appendix J) were significant, F(3,
103) = 1,07, p > 0,05. Hence, hypothesis 4c is rejected.
4.8 Type of disease
An Anova test was conducted to find an answer to hypothesis 5.
The descriptives showed no big differences between the means of the groups. If these
differences were significant was tested by the Anova with post hoc tests.
N Mean SE
chronic 53 3,53 ,131
medium term 17 3,71 ,166
once 35 3,50 ,143
The Anova test gave a p-value of 0,424, the means of the different groups were not different.
This implies that the post hoc tests (Appendix K) were not significant as well.
So, there was no significant difference between people who have drugs for chronic diseases,
people who have drugs for midterm or for one time, F(2, 102) = 0,87, p > 0,05. Hence,
hypothesis 5 is rejected.
4.9 Brand Loyalty, Price, Quality and Status
A factor analysis was used to explore which items influence attitudes towards generics. The
factors found were used in a multiple regression to find out whether their influence is
significant and if so, in which direction.
The factor analysis was conducted with Varimax rotation. All the outcomes of the factor
analysis and multiple regression can be found in Appendix L.
The factor analysis indicated 5 factors which are of influence on attitude towards generics:
„Awereness‟, „Quality‟, „Price Sensitivity‟, „Status‟ and „Safe‟. Although the variable „Safe‟ was
found, it was left out of the analysis. Besides, „Safe‟ contained two variables that were a
variable for other factors. The remaining two factors did not show a significant correlation.
It‟s remarkable that there‟s no factor for „Brand Loyalty‟. The variables for „Brand Loyalty‟
scored no eigenvalues above 1 and their correlation with other items was not significant,
therefore is was left out of the analysis.
Table 4.8.1: results of the Anova test
39
Master Thesis Stefan van der Goes
The factors which were found consisted out of the following variables:
Table 4.9.1: overview factors
When a factor consisted of at least 3 variables, Cronbach‟s alpha was used to test for
reliability. When a factor consisted of 2 variables, a correlation test was conducted.
Factor Cronbach’s alpha
Correlation (sig. level)
Awareness 0,8 Quality 0,72 Price Sensitivity 0,62 Status 0,5 (0,01)
Table 4.9.2: reliability of the factors
To test which of the variables of the factor analysis determine attitudes towards generics , a
multiple linear regression was conducted. A multiple linear regression takes the following
form:
Where the variables represent the following:
Yi = attitudes towards generic drugs X3 = Price Sensitivity
X1 = Awareness X4 = Status
X2 = Quality
13
The variable „I‟m sensitive for sales‟, was of influence for the factor „Status‟ too. Sensitivity for sales is sometimes seen when
there are high prices for status goods, as studied by Garretson, et al. (2002). The choice was made to take „I‟m sensitive for
sales‟ only into account for the factor „Price Sensitivity‟, as there is an obvious correlation between these two.
Factor consisted out of the following variables:
Awareness Awareness of the drug via others
Awareness of the drug
via commercials
Awareness of the drug
Quality To me, quality is most important when choosing a product or brand
When in doubt, I usually choose for
quality
When in doubt, I usually choose the
same brand as I always do
Price Sensitivity When in doubt, I usually choose the
cheapest brand
How important is the price when
buying non-prescribed drugs?
I’m sensitive for sales13
I do not get the drug reimbursed,
so I choose for the cheapest drug
Status When the price is high, that means the quality is high
Expensive brands reflects status
40
Master Thesis Stefan van der Goes
When including these variables in the regression, the regression had the following form:
β0Int+ β1Awareness + β2Quality+ β3Price Sensitivity+ β4Status + ε
The results of the regression analysis were as follows:
Variable B Std. Error P
(Constant) 3,675 ,597 ,000
Awareness ,119 ,082 ,149
Quality ,051 ,098 ,607
Price Sensitivity ,290 ,085 ,001
Status ,011 ,090 ,904
R²=0,114
Table 4.9.3: output regression analysis
The regression model, which was significant (ANOVA, p=0,004) showed that Price Sensitivity
is the only factor which determines attitudes towards generics. Price Sensitivity has a beta of
0,290, which implies that when one‟s price sensitivity rises with one unit (someone is price
sensitive) his/her attitudes towards generics rises with 0,290. In other words, one has a more
positive attitude towards generics when one is price sensitive.
All the other factors are non-significant, so they can not be interpretated.
Looking at the partial plots, there is no sign of heteroskedasticy. The residuals all have the
same variances.
There is no indication of too much correlation or multicollinearity. The scores for the
correlation coefficient are all below 0,8 and the collinearity diagnostics showed that none of
the VIF-scores was higher than 10.
The R² gives a number of 0,114, which means that 11,4% of the attitudes towards generics
can be explained with this regression model. To obtain a model that explains more about
attitudes towards generics, the regression model was extended with data from the previous
tests. In order to make this regression, some variables had to be computed into dummies.
Some tests consisted of dichotomous variables (i.e. gender, knowledge about generics,
use(d) generics, etc.) which were already dummies. For the categorical data, dummies were
made and put into the regression analysis. The dummy of the null hypothesis was used as
the control group and was therefore left out of the regression. To test for a gender effect, a
two-sided test was used, which implies that there wasn‟t a control group. As a control
41
Master Thesis Stefan van der Goes
variable „Female‟ was chosen, as this variable represents the majority of the respondents
(Field, 2005, p. 208). The results of the regression analysis were as follows:
Variable B Std. Error P
(Constant) 2,785 ,660 ,000
Male ,032 ,188 ,866
25-45 years ,303 ,242 ,213
45 years and older ,053 ,225 ,816
Urban -,425 ,192 ,029
Familiar with generics
Use(d) generics
-,238
-,214
,238
,218
,321
,328
SES: very low ,200 ,263 ,449
SES: low -,030 ,243 ,903
SES: high ,294 ,228 ,200
Frequency use of drug: medium -,113 ,278 ,686
Frequency use of drug: chronic -,124 ,216 ,567
Awareness ,061 ,088 ,490
Quality -,005 ,106 ,961
Price Sensitivity ,285 ,096 ,004
Status ,000 ,105 ,999
R²=0,201
Table 4.9.4: output regression analysis
This model shows an R² of 0,201, which means that 20,1% of the attitudes towards generics
are explained with this model. That is nearly 10% more than the previous model. The
regression model, which was significant (ANOVA, p=0,049) gives are similar results as the
outcomes of the previous tests.
Only Price Sensitivity and Urban are significant, both at 5%. Price Sensitivity shows a beta of
0,285, which is slightly less than the previous model. When the price sensitivity of someone
rises with one unit, their attitudes towards generics will rise with 0,285. Or, when people are
price sensitive, they are more positive about generics. That sounds logical because generics
are cheaper than brand name drugs.
The other significant factor is Urban. It shows a beta of -0,425. That implies that when
someone lives in an urban environment, he or she will be more negative about generics.
When „Urban‟ rises with one (as this is a dummy, it is 1 when someone lives in a urban
42
Master Thesis Stefan van der Goes
environment), their attitudes towards generics will decline with 0,425. So, they will be more
negative about generics. This is in line with the results from hypothesis 3. The results of
hypothesis 3 showed a significant negative relationship between urban and attitudes towards
generics. A similar effect can thus be found in this regression model.
To make sure the results of this regression analysis are reliable, it was checked if none of the
assumptions of the regression analysis (linearity, homoscedastticity, no multicollinearity)
were broken. Appendix L shows the plots of the residuals. None of the residuals showed
signs of heteroscedasticity or linearity. To check for multicollinearity, the VIF-scores were
examined. They should not be higher than 10, and none of them were. Furthermore, no
correlations with a value of 0,8 or higher were found, so there is no sign of multicollinearity.
What these results mean for brand name and generic drug manufactures will be discussed in
the next chapter.
4.10 Private labels
In the previous chapters the similarities and differences between generics and private labels
were discussed. It‟s interesting is to examine whether the factors that determine attitudes
towards generics determine attitudes towards private labels as well.
The factor analysis showed that awareness, price sensitivity, status and quality had influence
on attitudes towards generics. This was found for generics too. However, contrary to
generics, brand loyalty was a factor for private labels. Another factor found was
characteristics. This consisted of the variables that people use to compare private labels
against each other or against brand name products (price, quantity and quality). The next
table shows all the factors and the variables of which they consisted.
43
Master Thesis Stefan van der Goes
Factor consisted of the following variables:
Awareness Awareness of the product when grocery shopping
Awareness of the product by others
Awareness of the product via
commercials
Quality When in doubt, I usually choose for
quality
To me, quality is most important when choosing a product or brand
Expected quality of the grocery
Characteristics How important is price when buying
private labels?
The quantity of the product
compared to the A-brand
(expected) quality of the private label
How important is price when
grocery shopping?
Status Expensive brands reflects status
When the price is high, that means the quality is high
Price Sensitivity When in doubt, I usually choose the
cheapest brand
I’m sensitive for sales14
Brand Loyalty For my grocery shopping, I always
buy the same brands
I love to try new brands (reverse)
Table 4.10.1: overview factors
Several variables were left out of a factor, because they either had:
- a very low correlation with the other variables;
- a variable which was a variable for another factor too;
- had no relationship with the factor, regardless of the score.
The full Component Matrix can be found in Appendix M.
The reliability of the factors was tested with Chronbach‟s alpha. When a factor contained only
two variables, a (partial) correlation was used. The results of these tests can be found in the
table on the next page. The reliability tests can be found in Appendix N.
14
„I‟m sensitive for sales‟ was a variable for „Status‟ too. The choice was made to take only „I‟m sensitive for sales‟ into account
for the factor „Price Sensitivity‟, as there is an obvious correlation between these two.
44
Master Thesis Stefan van der Goes
Factor Cronbach’s alpha
Correlation (sig. level)
Awareness 0,87 Quality 0,76 Characteristics 0,8 Status 0,5 (0,01) Price Sensitivity 0,44 (0,01)
Brand Loyalty 0,38 (0,01)
Table 4.10.2: reliability of the factors
Before discussing these results in the next chapter, the table on the next page provides an
overview of all the tested hypotheses and the results.
45
Master Thesis Stefan van der Goes
Hypothesis Accepted/ Rejected
H1: The attitudes towards generics are
dependent of gender
Rejected
H2: Age has a negative effect on attitudes
towards generics
Rejected
H3: People in urban regions will have more
positive attitudes towards generics than people
living in rural or semi-rural areas
Rejected
H4a: Knowledge about generics has a positive
effect on attitudes towards generics
Rejected
H4b: Currently using generics has a positive
effect on attitudes towards generics
Rejected
H4c: People with a lower SES will have more
positive attitudes towards generics than people
with a higher SES do
Rejected
H5: People having a chronic disease will have a
more negative attitude towards generics than
people who don’t
Rejected
H6a: Brand Loyalty will have a negative effect
on attitudes towards generics
Rejected
H6b: A higher price sensitivity has a positive
effect on attitudes towards generics
Accepted
H6c: (perceived) Quality has a negative effect
on attitudes towards generics
Rejected
H6d: Status has a negative effect on attitudes
towards generics
Rejected
Table 4.10.2: outcomes tested hypotheses
46
Master Thesis Stefan van der Goes
5. Discussion
Both insurance companies and governments are trying to stimulate the use of generic
instead of brand name drugs, largely due to the rise of costs of medical expenditures. In the
previous chapter the hypotheses were tested, largely to get a segmentation of different
people and their attitudes towards generics. This chapter will discuss the results of this
research and give some recommendations. Furthermore, some recommendations for
policymakers and pharmacies are being made.
5.1 Demographics
Hypotheses 1- 5 were all concerned with relationships of demographics with attitudes
towards generic drugs.
Research on the type of environment showed that people living in a urban environment are
more negative towards generic drugs that people living in a semi-rural or rural environment.
This is remarkable, because when looking at the literature, people living in a urban
environment are most of the time (due to their lower social economic status) more price
sensitive. As generics are sometimes 30% of the price of a brand name drug, one could think
that this would positively influence their attitudes towards generic drugs. Furthermore, is was
found that those people had a higher percentage of chronic diseases. These means more
drugs and for a longer term. This are two causes for a raise in healthcare costs. And a rise in
healthcare costs is, especially for lower income groups, a reason to favour generics. Why the
opposite effect is found, isn‟t clear. A possible explanation is that they are more loyal to „their‟
drugs. However, brand loyalty was not found in the factor analysis, such an effect could not
be examined. This certainly is interesting for future research.
But not only accepted tests can be interesting, rejected hypotheses (from a t-test) can
provide information about attitudes towards generics as well. For instance, there is no
significant difference between men and women and their attitudes towards generics.
Furthermore, knowledge about generics doesn‟t make people more positive about generics.
This could be due to the fact that there nowadays are a lot generics (or private labels)
around. Therefore people could get used to these copycats.
5.2 Behavioral factors
Factor Analysis showed that the following behavioral factors could be of influence towards
one‟s attitudes towards generics: „Awareness‟, „Quality‟, „Price Sensitivity‟, „Status‟ and
„Safe‟. Linear multiple regression showed that only „Price Sensitivity‟ was significant.
The relation between „Price Sensitivity‟ and attitudes towards generics correspondents with
the results of other studies discussed in this study. When people are price sensitive, they are
47
Master Thesis Stefan van der Goes
more positive about generics. A lower price could convince someone to buy or accept the
generic drug, as Gartner and Kreling (2000) concluded. This works especially when people
perceive the generic drug as a risk. This financial incentive has to bigger when the perceived
risk is bigger. In other words, only when patients are convinced that the product is safe, the
price incentive plays a role.
The recommendations which arose out of the conclusions will be discussed in the next
paragraph.
Out of the different analysis came various conclusions. Some of this conclusions were
interesting for generic drug manufacturers, other for brand name drug manufacturers.
At first, the recommendations for generic drug manufacturers will be discussed, followed by
recommendations for brand name drug manufacturers.
5.3. Recommendations for generic drug manufacturers
As pharmacists have, like private labels, own brands for some cures (i.e. cough drops), some
of this research could be of interest to them. One of the results of this study was that people
in an urban environment have more negative attitudes towards generics. As price sensitivity
has an positive effect, it is recommended to pharmacies in urban regions to focus more on
the price of the generics. But only the lower price will not be enough to convince people
buying or accepting generics. It was to be very clear to them that they don‟t have a bigger
risk when using generics. So, when the quality is clearly communicated, the price advantage
is most clear to them.
Furthermore, the buying of private labels has a negative relationship with the number of
items on sale (Burton, et al., 1998). That‟s why pharmacies have to make sure that they don‟t
have any promotions, price reductions, coupon purchases, etc. for national brands when
trying to sell more non-prescribed generics.
5.4 Recommendations for brand name drug manufacturers
Only non-prescribed drugs can be marketed. Brand name drug manufacturers can
differentiate themselves in this market by being innovative. Moreover, a manufacturer of
more brand name drugs can add a brand with another name and target only price sensitive
customers. In this way one can still serve the market of quality seeking customers (lead by
price as a sign of quality) and the market for price sensitive customers.
Ingredient branding would be harder to implement in the drugs market, because a lot of
people are unaware of the working ingredient in their drug. People who are aware of the
working ingredient in drugs are people who work with them, i.e. general practitioners and
pharmacists. Manufacturers of brand name drugs should mainly focus on this market.
48
Master Thesis Stefan van der Goes
That this strategy pays off is shown by research of Osinga (2011). He found that pharmacists
and general practitioners usually stop to promote a drug when the marketing activities of that
drug are stopped (often when the patent is expired). A solution is to continue longer with the
marketing aimed towards pharmacists and general practitioners
As this research shows, price sensitivity is a factor in determining attitudes towards generics.
The more price sensitive one is, the more positive he or she is about generics. To determine
price sensitivity relative price differences are often used. By comparing two or more brands
of the same product, the relative price difference becomes clear. The bigger the difference
between the brand name and generic drug, the bigger the chance that the price sensitive
consumer will choose a generic drug. So, brand name drug companies should consider
lowering their prices, (to lower the relative price difference) in order to win back the price
sensitive consumer. So, focusing on price could be a way to win back the price sensitive
consumer and to compete with generics. This strategy could work for the short term, but a
danger is that generic manufacturers will lower their prices in response. To get in a price
competition with generic manufacturers is not an option, as that is a battle that cannot be
won. So, for the longer term the production of own generic brands, collaborating with generic
companies and/or targeting on general practitioners and pharmacists are better options.
5.5 Limitations and future research
This research has some limitations. First, the sample used for this study was too much
biased by some groups. For example, there were twice as much women in the sample as
men. Furthermore, the education groups were biased. The sample contained of 20% people
with a university degree (or studying at a university), which is too much compared official
numbers, where only 10% of the Dutch population has a university degree (CBS, Statistical
yearbook 2011, p. 176). Other groups where too small in the sample. There were only 8
people (6%) who are above 65 years. This percentage in the Netherlands is 16% (CBS,
Statistical yearbook 2011, p. 60) So, the results of the tests for age effects should be
interpreted with caution.
Secondly, the measurement of the Social Economic Status (SES) could be a limitation. The
indicators for this variable in this study were „Income‟ and „Education‟, whereas most other
studies used „Income‟, „Education‟ and „Occupation‟ as indicators for SES.
Third, for much tests the groups sizes were very different. For the testing of hypothesis 1 and
hypothesis 4b, one group was nearly twice as big as the other group. Non-parametric tests
were used too, but the results of both tests were the same.
49
Master Thesis Stefan van der Goes
Fourth, the questionnaire did not start with questions about generic drugs. First there was
asked about private labels. This method might be the best idea to educate people who are
unfamiliar with generics. On the other hand, it could blur some results. People who have a
prejudice towards private labels, could project this prejudice on generics when filling out the
questionnaire.
Fifth, the R-squared from the regression analysis showed a value of 0,233. Which means
that 23% of the attitudes towards generics are explained by the factors in the analysis.
Ideally, this percentage would be higher.
As was mentioned in the previous paragraph a more thorough study on people above 65
years and their attitudes towards generic drugs is recommended. This could be extended
with contacts with general practitioners, specialists and the frequency of drugs, compared to
other age groups. It would therefore be interesting to compare the frequency of using
(prescribed and non-prescribed) drugs and the attitudes towards generics for people above
65 years
This is one of the few studies on attitudes towards generic drugs in the Netherlands, so
further research is highly recommended. For instance, this study and most other studies are
focusing on pharmacists and general practitioners as the prescribers of drugs. It is interesting
to expand this to specialists in hospitals. They do prescribe drugs too and because of the
more serious diseases that are threatened in the hospital, people may be more negative
about generic drugs.
This study shows an negative effect for people living in an urban environment and their
attitudes towards generics. More research is needed on this subtopic as this group possess
various characteristics which indicate a more positive attitude towards generics.
As brand loyalty was not a factor which determines attitudes towards generics, it is
interesting to study whether brand loyalty for drugs actually exists. Maybe they are they just
searching for the best drug, regardless of the brand?
It was mentioned in the chapter previous research that the decrease in healthcare costs in
Sweden was largely due to the substitution of brand name drug for generics. An interesting
topic for further research would be to get more insight in this process. For instance, about
how much a brand name drug declines in price when a generic comes to the market.
50
Master Thesis Stefan van der Goes
Furthermore, effects as availability of other (generic or brand name) drugs for the disease
should be taken to account when studying this.
51
Master Thesis Stefan van der Goes
6. References
Adler, N.E., Boyce, T., Chesney, M.A., Cohen, S., Folkman, S., Kahn, R.L. & Syme, S.L.
(1994). Socioeconomic status and health. The challenge of the gradient. American
Psychologist, vol. 49, no. 1, 15-24.
Ailawadi, K.L., Neslin, S.A.& Gedenk, K. (2001). Pursuing the value-conscious consumer:
store brands versus national brand promotions. Journal of Marketing, 65, 71-89.
Andersson, K., Sonesson, C., Petzold, M.G., Carlsten, A. & Lönnroth, K. (2005). What are
the obstacles to generic substitution? As assessment of the behavior of prescribers,
patients and pharmacies during the first year of generic substitution in Sweden.
Pharmacoepidemiology and Drug Safety, 14, 341-348. doi:10.1002/pds.1055
Andersson, K., Bergström, G., Petzold, M. G. & Carlsten, A. (2007). Impact of a generic
substitution reform on patients‟ and society's expenditure for pharmaceuticals. Health
Policy, vol.81, no.2, 376-384.
Barwise, J. & Perry, J. (1981). Situations and attitudes. The Journal of Philosophy, vol. 78,
no. 11, 668-691.
Batra, R. & Sinha, I. (11th ed.) (2000). Consumer-level factors moderating the success of
private label brands. Journal of Retailing, vol. 76, no. 2, 175-191.
Bearden, W.O. & Mason, J.B. (1978). Consumer-perceived risk and attitudes toward
generically prescribed drugs. Journal of Applied Psychology, vol. 63, no. 6, 741-746.
Betaalbaar houden van geneesmiddelen (n.d.) Retrieved May 10, 2011, from
http://www.rijksoverheid.nl/onderwerpen/geneesmiddelen/
betaalbaar-houden-van-geneesmiddelen
Boguski, Mandl & Sukhatme (2009). Repurposing with a difference. Science, vol. 324, no.
5933, 1394-1394. doi: 10.1126/science.1169920
Borgherini, G. (2003). The bioqequivalence and therapeutic efficacy of generic versus brand-
name psychoactive drugs. Clinical Therapeutics, vol. 25, no. 6, 1578-1592.
Burton, S., Lichtenstein, D.R., Netemeyer, R.G. & Garretson, J.A. (1998). A scale for
measuring attitude toward private label products and an examination of Its
psychological and behavioral correlates. Journal of the Academy of Marketing
Science, vol.26, no. 4, 293-306.
52
Master Thesis Stefan van der Goes
Caspi, A. (1984). Contact hypothesis and inter-age attitudes: a field study of cross-age
contact. Social Psychology Quarterly, vol. 47, no.1, 74-80.
Central Bureau voor de Statistiek. (2011). Statistisch jaarboek 2011. Retrieved June, 24,
2011 from http://www.cbs.nl/NR/rdonlyres/0604BF87-AE22-451F-85EB-
7BCFBE22CF94/0/2011a26pub.pdf
Chaudhuri, A. & Holbrook, M.B. (2001). The chain of effects from brand trust and brand
affect to brand performance: the role of brand loyalty. Journal of Marketing, vol.65,
no.2, 81-93.
Chen, A. & Wu, S. (2008). Dispensing pattern of generic and brand-name drug in children.
Ambulatory Pediatrics, vol. 8, no. 3, 189-195.
Cunningham, I.C.M., Hardy, A.P. & Imperia, G.(1982). Generic brands versus national
brands and store brands. Journal of Advertising Research, vol.22, no.5, 25-32.
Doney, P.M. & Cannon, J.P. (1997). An examination of the nature of trust in buyer-seller
relationships. The Journal of Marketing, vol.61, no.2, 35-51.
Douven, R. & Meijer, A. (2008). Prijsvorming van generieke geneesmiddelen: forse
prijsdalingen in het nieuwe zorgstelsel. CPB Document, ISBN: 978-90-5833-383-4
Farley, J.U. (1964). Why does brand loyalty vary over products? Journal of Marketing
Research, vol. 1, no.4, 9-14.
Field, A. (2005). Discovering statistics using SPSS. London, United Kingdom: SAGE
Publications Ltd.
Fields (1975). Rural-urban migration, urban unemployment and underdevelopment, and job-
search activity in LDCs. Journal of Development Economics, 2, 165-187
Frisk, P., Rydberg, T., Carlsten, A. & Ekedahl, A. (2011). Patients‟ experiences with generic
substitution: a Swedish pharmacy survey. Journal of Pharmaceutical Health Services
Research, vol. 2, no. 1, 9-15. doi: 10.1111/j.1759-8893.2011.00036.x
Gartner, J.M. & Kreling, D.H. (2000). Consumer perceptions of risk and required cost savings
for generic prescription drug. Journal of the American Pharmacists Association, vol.
40, no. 3, article 406688. Retrieved January 08, 2010 from
http://www.medscape.com/viewarticle/406688
53
Master Thesis Stefan van der Goes
Gaither, C., Kirking, D.M., Ascione, F.J. & Welage, L.S. (2001). Consumers‟ view on generic
medications, Article… Retrieved Janaury 8, 2010, from,
http://www.medscape.com/viewarticle/414923 08-01-2010.
Garretson, J.A., Fisher, D. & Burton, S. (2002). Antecedents of private label attitude and
national brand promotion attitude: similarities and differences. Journal of Retailing,
78, 91-99.
Goldstein, J. (2010, January 8). Teva and the narrowing gap between generics and big
pharma. The Wall Street Journal Health Blog. Retrieved January 8, 2010, from
http://blogs.wsj.com/health/2010/01/08/
teva-and-the-narrowing-gap-between-generics-and-big-pharma/ 08-01-2010.
Grossman, G.M. & Shapiro, C. (1988). Foreign counterfeiting of status goods. Quarterly
Journal of Economics, vol. 103, no. 1, 79-100
Heikkilä, R., Mäntyselkä, P. & Ahonen, R. (2010). Do people regard cheaper medicines
effective? population survey on public opinion of generic substitution in Finland.
Pharmacoepidemiology and Drug Safety, vol. 20, no. 2, 185-191. doi:
10.1002/pds.2084
Hoch, S.J. & Banerji, S. (1993). When do private labels succeed? Sloan Management
Review, vol.34, no.4, 57-67.
Van Horen, Pieters & Stapel (2009). Unexpected benefits of being less rather than more
similar: the influence of consumer mindset and brand presence on copycat
evaluation. Advances in Consumer Research, vol. 36, 755-756
Jeuland, A.P. (1979). Brand choice inertia as one aspect of the notion of brand loyalty.
Management Science, vol. 25, no. 7, 671-682.
Keith, L.G., Oleszczuk, J.J., Stika, C.S. & Stine, S. (1998). Generics: what‟s in a name?
International Journal of Fertility and Women’s Medicine, vol. 43, no. 3, 139-149.
Kjoenniksen, I., Lindbaek, M., & Granas, A.G. (2006). Patients‟ attitudes towards and
experiences of generic drug substitution in Norway. Pharmacy World & Science, 28,
284-289.
Kotler, P. & Keller, K.L. (13th ed.) (2009). Marketing management. New Jersey: Pearson
Education Inc.
54
Master Thesis Stefan van der Goes
Loyd, B.H. & C. Gressard (1984). The effects of sex, age, and computer experience on
computer attitudes. Association for Educational Data Systems, vol. 18, no. 2, 67-77.
Lumpkin, J.R., Hawes, J.M. & Darden, W.R. (1986). Shopping patterns of the rural
consumer: exploring the relationship between shopping orientations and
outshopping. Journal of Business Research, vol. 14, no.1, 63-81.
Mackenbach, J.P., Martikainen, P., Looman, C.W.N., Dalstra, J.A.A., Kunst, A. & Lahelma,
E. (2005). The shape of the relationship between income and self-assessed health:
an international study. International Journal of Epidemiology, 34, 286-293
McManus, P., Birkett, D., Dudley, J. & Stevens, A. (2001). Impact of the minimum pricing
policy and introduction of brand (generic) substitution into the pharmaceutical benefits
schema in Australia. Pharmacoepidemiology & Drug Safety, vol.10, no.4, 295-300.
Meer mensen in stad dan op platteland. (April, 18, 2006). De Volkskrant
Morgan, R.M. & Hunt, S.D. (1994). The commitment-trust theory of relationship marketing.
Journal of Marketing, 58, 20-38.
Moyer, M.W. (2010). Drug problem. Women aren‟t properly represented in scientific studies.
Slate. Article 2261297. Retrieved July 9, 2011, from http://www.slate.com/id/2261297
Mills, D.E. (1999). Private labels and manufacturer counterstrategies. European Review of
Agricultural Economics, vol.26, no.2, 125-165
Ministerie van Volksgezondheid, Welzijn en Sport (n.d.) Geneesmiddelen.
http://www.rijksoverheid.nl/onderwerpen/geneesmiddelen 20-05-2011
Mott, D.A. & Cline, R.R. (2002). Exploring generic drug behavior, the role of prescribers
and pharmacists in the opportunity for generic drug use and generic substitution.
Medical Care, vol. 40, no. 8, 662-674.
Narasimhan, C. & Wilcox, R.T. (1998). Private labels and the channel relationship: a
cross‐category analysis. The Journal of Business, vol. 71, no. 4, 573-600.
Odin, Y., Odin, N. & Valette-Florence. P. (2001). Conceptual and operational aspects of
brand loyalty: an empirical investigation. Journal of Business Research, vol.53, no.2,
75-84
55
Master Thesis Stefan van der Goes
Oppenheim, A.N. (1992). Questionnaire design, interviewing and attitude measurement.
London: Pinter
Organisation for Economic Cooperation and Development. Future budget pressures arising
from spending on health and long-term care. Retrieved July 15, 2011, from
http://www.oecd.org/dataoecd/19/24/37740852.pdf
Organisation for Economic Cooperation and Development. Health expenditure and financing.
Retrieved July, 15, 2011, from
http://stats.oecd.org/index.aspx?DataSetCode=HEALTH_STAT
Osinga, E. (2011). Pharmaceutical marketing: its effects on drug sales and beyond,
Unpliblished manuscript. Rijksuniversiteit Groningen, 2011.
Paykal, E.S., Abbott, R., Jenkins, R., Brugha, T.S. & Melzer, H. (2000). Urban-rural mental
health differences in Great Britian: findings From the National Morbidity Survey.
Psychological Medicine, 30, 269-280.
Peen, J., Schroevers, R.A., Beekman, A.T. & Dekker, J. (2009). The current status of urban-
rural differences in psychiatric disorders. Acta Psychiatrica Scandinavia, vol. 121, no.
2, 84-93. doi: 10.1111/j.1600-0447.2009.01438.x
Podulka, M., Krautkramer, K. & Amerson, D. (1989). Consumers‟ attitudes toward generic
drugs. Journal of pharmaceutical marketing & management, 4 (1), 93-104.
Pol, E. & Carroll, P.G.H. (2006). An introduction to economics with emphasis on innovation.
Melbourne: Thomson.
Richardson, P., Jain, A.K. & Dick, A. (1996). The influence of store aesthetics on evaluation
of private labels. Journal of Product & Brand Management, vol. 5, no. 1, 19-28.
Rijksinstituut voor Volksgezondheid en Milieu. (2002). Sociaal-economische status en
toegankelijkheid van zorg in Nederland. Retrieved April, 21, 2011, from
http://rivm.openrepository.com/rivm/bitstream/10029/9177/1/279601002.pdf/
Rijksinstituut voor Volksgezondheid en Milieu (2010). Zijn er verschillen naar
sociaaleconomische status? Volksgezondheid toekomst verkenning. Retrieved July,
13, 2011, from http://www.nationaalkompas.nl/zorg/
genees-en-hulpmiddelen-lichaamsmaterialen/geneesmiddelen/
verschillen-sociaaleconomisch/
56
Master Thesis Stefan van der Goes
Rijksinstituut voor Volksgezondheid en Milieu (2006). Sociaaleconomische status 2006.
Volksgezondheid toekomst verkenning. Retrieved July 15, 2011 from
http://www.zorgatlas.nl/beinvloedende-factoren/sociale-omgeving/
ses/sociaaleconomische-status-2006/
Rosen, D.L. (1984). Consumer perceptions of quality for generic grocery products: a
comparison across product categories. Journal of Retailing, vol.60, no.40, 64-80.
Schneeweissab, S., Schöffskic, O. & Selked, G.W. (1998). What is Germany‟s experience on
reference based drug pricing and the etiology of adverse health outcomes or
substitution? Health Policy, vol. 44, no.3, 253-260.
Stichting Farmaceutische Kengetallen (2005) Anatomische verschillen [Electronic version].
Pharmaceutisch Weekblad, 46.
Suh, D.C. (1999). Trends of generic substitution in community pharmacies. Pharmacy World
& Science, vol. 21, no. 6, 260-265. doi:10.1023/A:1008781619011
Szymanski, D.M. & Busch, P.S. (1987). Identifying the generics-prone consumer: a meta-
analysis. Journal of Marketing Research, 24, 425-431.
Urbany, J.E., Dickson, P.R. & Kalapurakal, R. (1996). Price search in the grocery market.
Journal of Marketing, vol. 60, no.2 91-104.
Vaidyanathan, R. & Aggarwal, P. (2000). Strategic brand alliances: implications of ingredient
branding for national and private label brands. Journal of Product & Brand
Management, vol.9, no.9, 214-228.
Vaitheeswaran, V. (2009). Generically challenged. The pharmaceutical industry may find
inspiration in its old nemesis, the generic-drugs business. The Economist, Retrieved
June, 16, 2011, from http://www.economist.com/node/14742621/ 16-06-2011
Waber, R.L., Shiv, B., Carmon, Z. & Arielry, D. (2008). Commercial features of placebo and
therapeutic efficacy. Journal of the American Medical Association, vol 299, no.9 1016-
1017.
White, J.A. (2010, January 7). The latest deal in the big pharma-generics wars. The Wall
Street Journal Health Blog, Retrieved January 8, 2010, from
http://blogs.wsj.com/health/2010/01/07/
the-latest-deal-in-the-big-pharma-generics-wars/
57
Master Thesis Stefan van der Goes
Winkleby, M.A., Jatulis, D.E., Frank, E. & Fortmann, S.P. (1992). Socioeconomic status and
health: how education, income, and occupation contribute to risk factors for
cardiovascular disease. American Journal of Public Health, vol. 82, no. 6, 816-820.
Appendix A: Demographics
Gender:
N Percentage
Male 41 31
Female 91 69
Total 132 100
Age (groups):
N Percentage 0-15 0 0
15-25 32 25
25-45 44 34
45 years and older 52 41
Total 128 100
Area:
N Percentage
Urban 42 32
Semi-rural 18 14
Rural 71 54
Total 131 100
Income (euro’s):
N Percentage None 6 4
0-10.000 35 26
10.000-20.000 8 6
20.000-30.000 23 17
30.000-40.000 16 12
40.000-50.000 8 6
50.000 or more 17 13
Don’t want to say 23 17
Total 136 100
Educational level:
N Percentage Basisschool 4 3
VMBO/MAVO 10 7
MBO 23 17
HAVO 13 10
VWO 19 14
HBO 40 30
WO 27 20
Total 136 100
Appendix B: Questionnaire (in Dutch)
Appendix C: Overview of urban, semi-rural and rural
City
No. of inhabitants
km²
inhabitants per km²
definition
Alblasserdam 19.014 8,79 2163 urban Alkmaar 29.051 138,15 3206 urban Almere 188.160 129,81 1450 semi-rural Amersfoort 144.862 62,77 2308 urban Amstelveen 80.695 41,46 1946 urban
Amsterdam 767.457 165,92 4625 urban
Apeldoorn 15.5726 339,87 458 rural
Arnhem 147.018 97,94 1501 urban
Barendrecht 46.449 19,80 2346 urban
Bergen op Zoom 65.845 80,09 822 rural
Best 28.953 34,3 844 rural
Breda 173.299 126,63 1369 semi-rural
Capelle a/d Ijssel 65.345 14,26 4582 urban
Delft 96.760 22,83 4238 urban
Den Haag 488.553 81,88 5967 urban
De Bilt 42.017 66,32 634 rural
Duiven 25.593 33,89 755 rural
Eersel 18.157 82,41 220 rural
Eindhoven 213.809 87,70 2438 urban
Enschede 157.052 141,03 1114 semi-rural
Emmen 109.491 336,58 325 rural
Epe 32.881 156,19 211 rural
Gennep 17.153 47,81 359 rural
Goirle 34.663 18,92 1832 urban
Groningen 187.298 78,28 2393 urban
Haarlem 149.579 29,24 5116 urban
Heerenveen 43.418 135,07 321 rural
Heerhugowaard 51.178 38,36 1334 semi-rural
Heerlen 15.260 104,07 147 rural
Heiloo 22.451 18,7 1201 semi-rural
Hillegom 20.484 12,91 1587 urban
Hoogeveen 54.805 127,65 492 rural
Hoorn 70.252 20,23 3437 urban
Horst a/d Maas 41.465 188,6 220 rural
City
No. of
inhabitants
km²
inhabitants
per km²
definition
Leeuwarden 94.073 79,1 1189 semi-rural Leiderdorp 26.426 11,67 2264 urban Leidschendam 72.160 32,76 2203 urban Lelystad 74.628 231,72 322 rural Lochem 33.395 213,26 157 rural
Loppersom 10.452 111,03 94 rural
Maasdriel 23.756 66,02 360 rural
Maastricht 118.533 56,82 2086 urban
Middelburg 47.997 48,59 988 rural
Mill en St. Hubert 11.031 52,19 211 rural
Moerdijk 36.536 159,10 230 rural
Nijmegen 162.963 53,55 3043 urban
Noordwijk 15.555 22,59 689 rural
Noordoostpolder 46.090 460,32 100 rural
Oegstgeest 22.597 7,18 3147 urban
Oldampt 39.486 227,85 173 rural
Oldebroek 22.750 97,73 233 rural
Oirschot 17.750 101,81 174 rural
Oudekerk 8.151 27,03 302 rural
Papendrecht 31.853 9,46 3367 urban
Purmerend 79.038 23,44 3372 urban
Ridderkerk 44.746 23,72 1886 urban
Roermond 55.212 60,91 906 rural
Roosendaal 77.566 106,51 728 rural
Rotterdam 605.543 208,84 2900 urban
Schiedam 75.565 18,03 4191 urban
Schinnen 13.494 24,08 560 rural
Schoonhoven 11.985 6,27 1911 urban
Sliedrecht 24.051 12,81 1878 urban
Smallingerland 55.271 118,28 467 rural
Steenwijkerland 43.208 290,3 149 rural
Strijen 8.916 51,13 174 rural
Terneuzen 54.878 251,15 219 rural
Terschelling 4.733 87,08 54 rural
Tilburg 204.853 117,32 1746 urban
Utrecht 307.081 94,65 3244 urban
City
No. of
inhabitants
km²
inhabitants per km²
definition
Vianen 19.647 39,43 498 rural
Voorst 23.772 123,1 193 rural
Wassenaar 25.816 50,96 507 rural West Maas en Waal 18.413 77,40 238 rural Westland 99.717 79,52 1254 semi-rural Winterswijk 29.051 138,15 210 rural Wormerland 15.862 38,74 409 rural Zaanstad 145.332 73,87 1967 urban
Zaltbommel 26.428 79,72 332 rural
Zoetermeer 121.532 34,56 3517 urban
Zoeterwoude 8.118 21,21 383 rural
Zwolle 119.030 111,37 1069 semi-rural
Appendix D: Policy regarding generic drugs (in Dutch)
Betaalbaar houden van geneesmiddelen Het ministerie van VWS maakt beleid om de kosten van geneesmiddelen te beheersen. Anders zouden de prijzen van medicijnen elk jaar minimaal 10% stijgen. Om medicatie betaalbaar te houden, heeft het ministerie een aantal maatregelen getroffen. Nieuwe geneesmiddelen kritisch toelaten Zorgverzekeraars mogen niet vanzelfsprekend alle nieuwe geneesmiddelen vergoeden. De minister van VWS besluit welke nieuwe medicatie toegelaten mag worden tot het verzekeringspakket. Het beleid is om kritisch toe te laten, zodat het verzekeringspakket niet buitensporig groot wordt. Lagere prijzen doorberekenen Apothekers berekenen sinds enkele jaren lagere prijzen door aan patiënten en verzekeraars. Dit is het gevolg van afspraken tussen het ministerie van VWS, apothekers, fabrikanten van merkloze medicijnen en zorgverzekeraars. Maximumprijzen Het ministerie van VWS kan maximumprijzen vaststellen voor geneesmiddelen vanwege de externe link: Wet Geneesmiddelenprijzen. De prijzen in de landen om ons heen gelden hierbij als richtlijn. Vóór de invoering van deze wet lagen de geneesmiddelprijzen in ons land 20% hoger dan in omringende landen. Bijbetalen in de apotheek Voor bepaalde geneesmiddelen geldt dat een patiënt moet bijbetalen wanneer de prijs boven de limiet ligt uit het externe link: Geneesmiddelenvergoedingssysteem (GVS). Patiënten merken hier meestal niets van, omdat de arts al rekening houdt met het GVS bij het uitschrijven van het recept. Zelfde werking Het GVS bevat een lijst van geneesmiddelen die in grote lijnen dezelfde werking hebben. Dat kunnen middelen zijn met dezelfde werkzame stof, maar ook middelen waarvan de werkzame stof verschillend is. Het gaat erom dat de middelen hetzelfde effect hebben op de aandoening van de patiënt. Het GVS groepeert deze geneesmiddelen in clusters. Per cluster geldt een maximale prijs die vergoed mag worden. Artsen zullen meestal niet de duurste recepten voorschrijven uit een cluster, omdat het effect van de middelen hetzelfde is. Het ministerie van VWS bepaalt de inhoud van het GVS op basis van adviezen van het externe link: College voor zorgverzekeringen (CVZ). Goedkoopste variant vergoed Het is mogelijk dat u bij de apotheek een ander doosje krijgt dan u gewend bent. Het doosje en het medicijn lijken anders, maar de apotheker verzekert u dat de werking hetzelfde is. Dat komt omdat de meeste zorgverzekeraars sinds 1 juli 2008 alleen de goedkoopste variant vergoeden van medicijnen met dezelfde werkzame stof. Dit heet het preferentiebeleid. Elke zorgverzekeraar bepaalt zelf welke variant zij vergoedt. Het preferentiebeleid levert financieel voordeel op voor de zorgverzekeraar en voor patiënt. Door goedkopere varianten van hetzelfde medicijn te gebruiken:
verminderen de uitgaven aan medicijnen door zorgverzekeraars;
zullen de ziektekostenpremies voor de patiënt minder hard stijgen;
betaalt de patiënt de laagste prijs wanneer het medicijn onder het ‘eigen risico’ valt binnen de zorgverzekering.
Duurdere variant soms vergoed Het preferentiebeleid betekent niet dat een arts automatisch het goedkoopste medicijn voorschrijft. Een arts kan op een recept vermelden dat een andere, duurdere variant van hetzelfde medicijn, noodzakelijk is voor een patiënt. De apotheek verstrekt dan de duurdere variant. De zorgverzekeraar vergoedt dit medicijn als het is opgenomen in het basispakket van de zorgverzekering. Elektronisch voorschrijfsysteem: betere recepten Artsen schrijver beter en minder recepten voor wanneer zij gebruik maken van het elektronisch voorschrijfsysteem (EVS). Dit systeem controleert de recepten die artsen invoeren op hun computer en stuurt deze automatisch door naar de apotheek. Artsen maken hierdoor minder fouten, bijvoorbeeld in het bepalen van de juiste dosis. En apothekers ontvangen altijd een volledig ingevuld en leesbaar recept. De overheid en de koepels van artsen en apothekers proberen het ‘zinnig en zuinig’ voorschrijven te bevorderen onder artsen. Vergoeding van dure geneesmiddelen Ziekenhuizen gaan vanaf 2011 onderhandelen met zorgverzekeraars over de kosten voor bepaalde dure geneesmiddelen. Dat is het gevolg van de invoering van prestatiebekostiging in 2011. Hierdoor moeten deze dure geneesmiddelen betaalbaar blijven. (taken from: http://www.rijksoverheid.nl/onderwerpen/geneesmiddelen/betaalbaar-houden-van-geneesmiddelen#anker-lagere-prijzen-doorberekenen)
Appendix E: Output Hypothesis 1
Appendix F: Output Hypothesis 2
Appendix G: Output Hypothesis 3
Appendix H: Output Hypothesis 4a
Appendix I: Output Hypothesis 4b
Appendix J: Output Hypothesis 4c
Appendix K: Output Hypothesis 5
Appendix L: Output hypothesis 6a, 6b, 6c and 6d (Factor Analysis and Multiple Linear Regression)
Reliability analysis for ‘Awareness’
Reliability analysis for ‘Quality’
Reliability analysis for ‘Price’
Correlations for ‘Status’
Output Regression analysis
Appendix M: Output overall regression
Appendix N: Factor Analysis Private Labels